Search

The attenuation of low-frequency sound in the sea increases with pH. In most ocean areas, the value of pH changes significantly with depth, and must be included in an accurate attenuation computation. Previous determinations of other parameters, such as surface loss, which were based on older attenuation formulas, are therefore inaccurate. An analysis of a previously reported surface loss formula indicates that the predicted values are too high. >

Nutritional aspects of health care for women are receiving special attention in the lay and research press. Primary care physicians must have accurate knowledge of general nutrition to counsel patients accordingly. This article addresses a practical approach to nutritional assessment and counseling. Current data on specific nutrients and disease are evaluated. Attention is given to antioxidant vitamins and disease as well as calcium, vitamin D, and osteoporosis.

Varicella may be associated with serious complications including encephalitis, Reye's syndrome, and drug toxicity. In this case, a 19-month-old child with varicella was brought to the family practice clinic by her parents when she began behaving abnormally. At the time of presentation the child exhibited dilated pupils, ataxia, urinary retention, and facial grimacing. The child's parents had treated her with acetaminophen, diphenhydramine syrup, colloidal oatmeal baths, and frequent applications of Caladryl lotion. The results of her immediate laboratory tests were within normal limits, and she was admitted to the hospital for observation. She recovered without therapeutic intervention. Although not available at the time of admission to the hospital, her diphenhydramine serum level was 1948 ng/mL. Diphenhydramine levels above 100 ng/mL have been associated with toxicity.

A series of propagation loss measurements over a single surface reflected path is analyzed using the pH‐dependent global attenuation model. Values of surface loss for the frequency range 10–2000 Hz were obtained at low sea state conditions. The results for incident angles between 10° and 24° were found to be very small and in general agreement with the work of Adlington [J. Acoust. Soc. Am. 35, 1834–1835 (1963)]. A comparison is made to presently used surface loss models. [Work supported by NUSC.]

Expendable bathythermograph (XBT) and salinity observations were made along a north‐south section across the South Fiji Basin in March 1976, and XBT observations only were made in May 1976, January‐February 1977, and July 1978. The salinity section showed a tongue of low salinity (less than 34.75%c) surface water associated with the intertropical convergence zone of winds, near 20°S. The corresponding temperature section indicated the presence of weak eddies; a warm‐core eddy near 28°S, 176°E was encountered in February 1977. The main feature of the subsurface temperature structure was a thermal front between 25CS and 26°S at depths between 100 and 400 m: the temperature at 200 m was over 18°C north of the front and below 18°C to the south of it. This front, observed on all the temperature sections, was strongest from January to March. It is ascribed to the flow of a subtropical counter‐current similar to that encountered in the South Indian Ocean at about the same latitude.

A long‐range sound propagation experiment conducted in the Subarctic transition zone (40°–50° N) in the Northeast Pacific Ocean has permitted the investigation of the anomalously large attenuation values previously reported. Our results, based on an experimental track with relatively uniform propagation conditions in the zone (parallel to the boundaries), support ‘‘normal’’ attenuation values. This suggests that the earlier anomalously high values may have been an artifact caused by rapidly changing propagation conditions along the path selected (traversing the zone).

An earthquake of magnitude 6⋅9 Ms occurred on March 24 1976 with its epicentre on the Kermadec Ridge. Shortly after the event large underwater acoustic signals were recorded on a hydrophone suspended in midwater over the South Fiji Basin. The travel times of the signals are related to the crustal structure along the path between the epicentre and the hydrophone. Reverberations from bathymetric features surrounding the South Fiji Basin were observed for over half an hour after the main arrivals and signals attributed to aftershocks were recorded for at least 2½ hours afterward.

Low‐frequency propagation measurements in sound channels often show a frequency‐independent excess loss that is evidently caused by internal scatter from large‐scale inhomogeneities. Using the Garrett–Munk internal wave spectrum (GM 75) as the scattering model, we have obtained an approximate solution of the ray‐diffusion equation on the deep sound‐channel axis. The estimated attenuation coefficient, ∠5×10−4 dB/km, is consistent with the smaller experimental values reported.Subject Classification: [43]30.20; [43]20.20; [43]30.40.

The deep ocean sound channel is used to obtain very long range (typically >2,000 km) acoustic transmission1 via totally refracted propagation paths (SOFAR propagation2). Such experiments can therefore determine the acoustic transmission properties of large areas of ocean3. Those acoustic propagation experiments can be used to locate major ocanographic changes4 and to identify specific water masses5. A 10,000-km underwater sound transmission experiment conducted between New Zealand and Peru to obtain an acoustic cross-section of the South Pacific Ocean is described here. Three distinct regions of transmission loss were found. The highest attenuation, which is attributed to Antarctic intermediate water, occurred in the central South Pacific Ocean.

Two surface duct propagation experiments are examined in which absorption and diffraction appear to be the dominant attenuation mechanisms. The absence of an excess diffusion loss suggests that little or no internal scatter occurs in a mixed layer. This result is consistent with the internal wave scatter hypothesis.

A number of experiments carried out over the past two decades tend to confirm the existence of two major anomalies at infrasonic frequencies, (i) Long-range propagation measurements in the sound channel show extra attenuation that cannot be the result of medium absorption (ii) Vertical directionality measurements of ambient noise show a broad maximum in the axial direction where a minimum is expected. A possible mechanism for both phenomena is scattering by internal waves or eddies.

This document presents the oral and visual presentation entitled 'A Further Study of the Space and Time Stability of a Narrowband Acoustic Signal in the Ocean: Short Range Results,' presented at the 102nd Meeting of the Acoustical Society of America, 1 December 1981, in Miami Beach, Florida. In the last of a three paper series on the space and time stability of a narrowband acoustic signal, results from short (16-17 nmi) ranges are presented. In contrast to the data obtained at longer ranges (P. D. Herstein, et al., J. Acoust. Soc. Am. 69 (S1), S33(A), 1981), an increased frequency spread was found in the received signal due to the greater number of possible raypaths at short ranges. A marked change in the character of the received signal was observed at ranges less than 70 nmi. Space and time variability are given for percentage Doppler shift and bandwidth. Comparison is made to data reported in the two previous papers and with predictions of the Multipath Expansion Option of the Generic Sonar Model.

Braces can be divided according to their use: rehabilitative, functional, or prophylactic. Rehabilitative braces are used following injury or surgery to allow protection of the injured part while rehabilitation is taking place. Functional braces are used following injury or surgery to allow an individual player to return to athletic activities while protecting the injured part from further damage. Prophylactic braces are used in an attempt to prevent injury to a normal area while engaged in athletic endeavors. Braces are available for the neck, back, and both the upper and lower extremities. There are many commercially available braces that are quite useful and effective. It is still controversial whether or not prophylactic braces prevent knee injuries in contact sports. However, prophylactic braces and/or taping have proved to be quite effective in preventing ankle injuries. It should always be kept in mind, however, whether one is using rehabilitative, functional, or prophylactic braces, that a good rehabilitation program and conditioning should be instituted along with their use to strengthen the involved area and prevent further injuries.

Acoustic propagation loss is a critical factor in the design and performance of sonar systems. Absorption by the medium is generally the dominant attenuation mechanism in the sonar equation and is the subject of this paper.

The water mass regions in the North Pacific are relatively well defined. Kibblewhite et al. [J. Acoust. Soc. Am. 61, 1169–1177 (1977)] have shown that each water mass has a characteristic low frequency (below 200 Hz) attenuation. They find the highest attenuation to be in the Subarctic‐Subtropical Transition Zone and these values are the highest yet reported in any ocean for this frequency range. It has recently been reported by Focke et al. [J. Acoust. Soc. Am. 71, 1438–1444 (1982)] that such values of attenuation can be modeled using a frequency independent mechanism, exponentially decreasing from the surface, the origin of which is unspecified. We compare these results with attenuation coefficients we have obtained in this zone. The oceanographic and biological properties of this region are analyzed to determine possible scattering mechanisms and their distribution. The effect of the two frontal areas bounding this region on the determination of attenuation coefficients is also examined.

UNDERWATER sound propagation measurements in natural bodies of water frequently reveal attenuation anomalies that cannot be ascribed to known processes. When chemical relaxations are suspected, the acoustic resonator–decay method has proved an effective means of investigation in laboratory conditions. By stepwise synthesis of an artificial medium of similar composition and measuring changes in decay rates, the responsible chemical constituents can be identified. Then by combined stoichiometry and acoustic measurements with different concentrations, the relaxation kinetics of the reacting species can be determined. We have shown previously by resonator measurements that the acoustic absorption in Lake Tanganyika is the result of a chemical relaxation involving magnesium and carbonic acid1. The same relaxation is also believed to be responsible for one of the absorption components in seawater2. We show here that the relaxation kinetics follow a two-step association of the magnesium carbonate ion-pair.

Because of its inherent stratification, the ocean can be modeled by a canonical sound speed profile which determines an idealized, range‐independent waveguide from which the first‐order acoustic propagation characteristics can be computed [W. H. Munk, J. Acoust. Soc. Am. 55, 220 (1974)]. Perturbing this stratification on the mesoscale produces a canonical eddy characterized by a first baroclinic mode vertical structure and a Rossby deformation radius horizontal length scale. This parameterization enables the acoustic effects of ocean eddies to be estimated using various range‐dependent propagation models—at least as far as first‐order characteristics are concerned. [Work supported by NUSC.]

In the last of a three paper series on the space and time stability of a narrow‐band acoustic signal, results from short (15–50 km) ranges are presented. In contrast to the data obtained at longer ranges [Herstein et al., J. Acoust. Soc. Am. Suppl. 1 69, S33 (1981)] an increased frequency spread was found in the received signal due to the greater number of possible raypaths at short ranges. A marked change in the character of the received signal was observed at ranges less than 30 km. Space and time variability are given for percentage Doppler shift, bandwidth, and intensity. Comparison is made to data reported in the two previous papers and with predictions of the Multipath Expansion Option of the Generic Sonar Model. [Work supported by DARPA and ELEX PME 124.]

In a previous paper (D. G. Browning et al J. Acoust. Soc. Am. 61, S10(A) (1977)) a large oceanographic feature, designated as a swirl, was identified in the Western Mediterranean Sea. Acoustic propagation loss was predicted based on archival environmental data using both single and multi‐profile models. These results are now compared with improved predictions using detailed environmental data and with corresponding experimentally measured propagation loss. It is found that the change in sound speed structure across the swirl significantly effects the propagation of sound. (Work supported by NUSC)

Low frequency propagation measurements in sound channels often show a frequency‐independent attentuation that is evidently caused by scatter from internal inhomogeneities. Using the Garrett‐Munk internal wave model [J. Geophys. Res. 80, 291–297 (1975)] as the scatter mechanism, approximate solutions for the ray‐diffusion equation are obtained for typical inhomogeneous sound channels with source and receiver on the axis. The resulting attentuation coefficients are in good agreement with experiment. [Work supported by NUSC.]

The Straits of Bab‐el‐Mandeb is a natural constriction between the highly saline Red Sea and the Gulf of Aden (Indian Ocean). The merging of these two distinct water masses in the Straits results in a highly stratified sound‐speed profile consisting of a high sound‐speed layer overlying a low speed layer. A sound propagation experiment utilizing shots detonated at 18 m was conducted across the Straits along a relatively flat sandy bottom with an average water depth of 60 m. Results are compared with modeling predictions using the Fast Field Program (FFP).

A persistent large (200‐km‐diam) warm core eddy has been reported [S. Tabata, J. Phys. Oceanogr. 12, 1260–1282)] in the Northeast Pacific Ocean off Sitka, Alaska (centered at 57° N Lat., 139° W. Long.). Not associated with a strong boundary current, this eddy differs from a previously observed South Pacific warm‐core eddy [P. Scully‐Power et al., J. Acoust. Soc. Am. 63, 1381–1388(1978)]. It appears to be formed from an influx of warm water at 100–500 m which moves northward along the North American Coast. At these high latitudes the surrounding water mass is relatively cold with a deep sound channel axis at 200 m and a secondary channel axis at 100 m. It was found, as in other warm core eddys, that there is a broadening of the secondary sound channel. For this eddy there is also a significant change in the depth of the deep sound channel axis (400 m at center of eddy, 200 m outside the eddy). An analysis was conducted using a range dependent propagation modeling program for a sound speed cross‐section of the eddy. Results which show the relative effect on sound propagation are given for various source and receiver depths. Computations are also made of the frequency dependence of both the deep and secondary sound channels and their interdependence as suggested by Hall [M. Hall, J. Acoust. Soc. Am. 66, 1102–1107 (1979)].

The axial arrival pattern at low frequencies (below 100 Hz) and relatively long ranges (200 km) from a sound channel propagation experiment in Baffin Bay shows an extensive series of arrivals after the refracted arrival pattern. Since the bottom at this location was relatively lossless, the hypothesis that these later arrivals represented energy that was diffracted from the sound channel was examined using the Fast Field (Modeling) Program. A comparison of sound‐propagation characteristics was made for the frequency range 1–100 Hz, assuming both high and low loss bottoms. The results show that diffraction does not significantly change the average propagation loss at frequencies above 5 Hz.

It is well known that low frequency sound waves can travel long distances in the ocean via the deep ocean channel (SOFAR propagation) and thus sample a large region. At low frequencies two attenuation mechanisms are dominant; a chemical absorption process due to boron, and scattering from oceanographic inhomogeneities. Hence attenuation can be related to the physical and chemical properties of the medium, and a characteristic attenuation can be attributed to a specific water mass. Data from long range propagation experiments in the Pacific Ocean is analyzed to determine the extent of particular water masses and to locate their boundaries (fronts).

A study is made of the stability of the principal parameters of a narrow‐band acoustic signal transmitted over a long range to a field of receivers. Measurement is made of the center frequency, bandwidth, and peak intensity as a function of both space and time. The relative stability of each parameter is determined and the possibility of identifying a single disturbance event is discussed.

Above 45 °N latitude the circulation of the Northeast Pacific Ocean is controlled by the counterclockwise Alaskan Gyre of subarctic water. Due to a greater freshwater influx than evaporation, the water column is characterized by the three distinct layers: 0–100 m depth, cold, low salinity water; 100–200 m, zone of strong salinity gradient (halocline); 200 m and below, gradual changes in temperature and salinity. Roden [J. Geophys. Res. 69, 2899–2914 (1964)] has shown that the middle halocline layer is the site of a complex temperature structure with numerous temperature inversions. These temperature inversions result in the formation of secondary sound channels which our analysis shows are widely distributed. Using a PE prediction model the cutoff frequency of a typical secondary sound channel was determined to be approximately 80 Hz. Since depth excess exists at most locations, energy not trapped in the secondary duct remains in the deep sound channel. Relative loss in the two channels is presented as a ...

Coastal ocean upwellings occur throughout the world's seas; many are regions of high marine productivity. Two factors associated with these features could significantly effect sound propagation: Abrupt temperature changes and a sloping topography. An analysis is made utilizing a range dependent propagation model and environmental data from upwelling areas off the coast of California and Brazil. The effect on sound propagation is compared with adjacent open ocean conditions for various source and receiver depths over a frequency range of 10–5000 Hz.

There are three principal causes of shallow or secondary sound channels in the North Pacific Ocean: Temperature inversions, water mass intrusions, and dynamic interleaving. Our analysis shows temperature inversions to be widespread throughout the Subarctic North Pacific, water mass intrusions and dynamic interleaving are more associated with specific locations. Examples are given of each type of shallow sound channel, showing their acoustic and oceanographic characteristics. Estimates of the distribution and duration of these channels is given.

An analysis of two extensive sound speed profile surveys (winter and summer) of the subarctic region (above 40° N) in the Northeast Pacific Ocean shows that the depth of the deep sound channel axis is symmetric about the Alaskan Gyre (centered at 53° N, 153° W). The axis depth ranges from 100 m or less in the Gyre to 400 m or more along the coast of North America. A model for axis depth contours is developed which has some interesting results. For example, a north‐south track along 140° W has little change of axis depth with latitude, while an east‐west track along 53° N would have a significant change in axis depth. An estimate of the yearly variability of the model is made from long‐term oceanographic data obtained at weather station “Papa” (50° N, 145° W). Sound propagation loss is computed for selected tracks using a range‐dependent acoustic prediction model and these results are compared to experimental data.

Previous long range acoustic experiments show that the Mid‐Atlantic Ridge, a major topographic feature rising to the deep sound channel axis, can have a significant effect on SOFAR propagation. [R. J. Urick, J. Acoust. Soc. Am. 35, 1413 (1963)]. In order to quantify this effect, data has been analyzed from a recent SOFAR experiment which deployed SUS charges during several transits across the Ridge. The signals were received on a hydrophone located near Bermuda, a distance of approximately 2500 km. These results are compared with data from Atlantic seamounts of similar height and ridges in other oceans. [K. M. Guthrie, J. Acoust. Soc. Am. Suppl. 1 68, S52(A) (1980)]. The enhancement or shadowing of SOFAR propagation is presented as a function of source depth and frequency for various geometries. [Work supported by NAVSEA.]

A series of long range propagation runs were analyzed to determine attenuation coefficients for the frequency range 25–1000 Hz. Above 200 Hz, the dominant cause of attenuation was boron absorption; below 200 Hz, the attenuation was independent of frequency indicating a scattering mechanism as the cause. In the Labrador Basin, absorption was close to that predicted by the Thorp formula; in the Davis Strait and Baffin Bay, it was significantly higher (1.3–1.5 Thorp). This increase, corresponding to higher pH values, can be attributed to two possible causes, the influence of the oxygen‐rich Labrador current and the higher pH values encountered as the sound channel axis shallows. The frequency independent component was similar in the Labrador Sea and Davis Strait but lower in Baffin Bay confirming results reported earlier [R. H. Mellen et al., J. Acoust. Soc. Am. 57, 1201–1202(L) (1975)] and suggesting less scattering in Baffin Bay. (Work supported by NUSC)

During Project LADLE a series of large explosive charges were detonated in the ocean near the Puerto Rico trench. Signals were received on hydrophones and a seismograph at a range of approximately 1500 km. An analysis is made of travel times and sound intensities. The results show the relative importance of water‐borne, water‐borne/seismic, and seismic transmission paths.

It is known that sound propagation in the ocean can be highly variable with time; however, it has not been determined if this variability is significant under typical sonar propagation conditions. To answer this question, a controlled experiment was conducted in the Tongue of the Ocean, Bahama Islands. A shallow source was suspended at various ranges from a fixed hydrophone array and a time series of transmission loss measurements obtained. A computer program (designated STAMP—Statistics of Acoustic Measurements and Prediction) was utilized to analyze these reults. Statistics are compared for both peak and energy propagation loss analysis. A comparison of the variability of direct path and bottom reflected arrivals is also made. [Work supported by NUSC.]

In the South Atlantic Ocean the cold Malvinas current flowing up from the South meets the warm Brazil current flowing down from the North off the coast of Argentina near Buenos Aires. The confluence of subtropical and subantarctic water masses results in a complex oceanographic regime. As a basis for a future measurement program, acoustic propagation predictions have been made for the region utilizing raywave and other modeling programs. Results are compared to a previous experiment conducted through the subtropical convergence near New Zealand [R. W. Bannister et al., J. Acoust. Soc. Am. 62, 847–859 (1977)].

A low frequency (50–800 Hz) sound‐propagation experiment was conducted along a 1400‐km path running eastward from Bermuda toward the Mid‐Atlantic Ridge. SUS charges were detonated at depths of 18, 154, 615, and 1230 m. The receiver was located at the axis of the deep sound channel (1250 m) at a range of approximately 1300 km. The acoustic path crossed several seamounts of the Corner Seamount Group. The highest of these peaks rose to the sound axis. This paper presents the relative enhancement of signal level for SOFAR propagation due to these seamounts as a function of source depth and frequency. The enhancement was minimal for the 1230‐m shots, while the greatest enhancement occurred for the 18‐m shots at the 50‐Hz filter band. This implies these seamounts and other topographic features such as the Mid‐Atlantic Ridge can significantly increase the coupling of low‐frequency ship‐generated noise into the deep sound channel. [Work supported by NAVSEA.]

Two low frequency attenuation mechanisms have recently been identified; a boron relaxation process and scattering from oceanographic inhomogeneities. It is now possible to obtain at least semiquantitative predictions of attenuation of sound in the sea for all frequencies of practical interest. Unfortunately, the available sonar slide rules are based on out dated or incomplete attenuation formulas. This paper suggests new engineering formulas based on present at‐sea and laboratory data applied to realistic propagation situations. [Work partially supported by Naval Underwater Systems Center.]

In a previous paper [A.D. Cobb et al., J. Acoust. Soc. Am. 67, S30 (1980)] a comparison was made between water‐borne and seismic arrivals from large explosive charges during Project LADLE. This paper presents a detailed analysis of the seismic arrivals and a comparison with data obtained from the geophysical investigations. In the 0–10 Hz filter band two types of principal seismic paths were found: an oceanic crust travel path (average sound speed 5.66 km/s) was obtained from three shots; an upper mantel path (average sound speed 7.82 km/s) was obtained from four shots. The seismic arrivals received on the bottomed hydrophones were approximately 100 dB lower in level than the water‐borne acoustic arrivals.

Low‐frequency ocean ambient noise data, when not dominated by shipping noise, show evidence for wind‐dependent noise at frequencies less than 500 Hz. Vertical directionality measurements have a horizontal component with a broad frequency characteristic. This effect is partly due to the coupling of wind‐generated noise into the sound channel by either a shallowing sound channel or a down‐slope conversion process due to basin boundaries and sea mounts. Omnidirectional measurements made below the sound channel critical depth, in sparsely shipped basins, and at high sea states indicate two distinct regions divided by the occurrence of breaking waves. Prior to wave breaking, a possible sound generation mechanism is the interaction of surface waves and turbulence in the near‐surface layer [I. Z. V. Gonchavov, Atmos. Oceanic Phys. 6(11) (1970); Yen and Pertone NUSC TR5833 (1979)]. Wave breaking produces bubble clouds. The dynamic evolution of these bubble clouds is a mechanism for the production of sound of frequencies less than 500 Hz. These clouds of micron‐size bubbles are regions of low sonic velocity described by Wood's volume fraction equations. These regions can be treated as a compressible body with a composite mixture speed and density that can exhibit a collective resonant oscillation and radiate as monopole and dipole sources. However, due to the proximity of the sea surface, only the monopole and its image, an effective dipole, would be of importance. When driven by the wave breaking vorticity and turbulence, these regions are shown to result in sufficient radiated sound to produce noise levels comparable to those observed and are also pronounced scatterers of low‐frequency sound.

In a previous paper [J. T. Malay et al., “Modeling sound propagation through a coastal ocean upwelling,” J. Acoust. Soc. Am. Suppl. 1 68, S78 (1980)] modeling predictions were made for the effect of a coastal upwelling on sound propagation. To verify these results, a sound propagation experiment was conducted across a moderately developed upwelling located at the Sur Canyon off the coast of California. A 1‐kHz projector was towed through a sonobuoy field which spanned the upwelling region. A supporting oceanographic survey was made to obtain environmental data input for multi‐profile modeling predictions. The relative importance of oceanographic and topographic changes on sound propagation were determined. A follow‐on experiment in a fully developed upwelling is planned. [Work supported by NUSC.]

Sound‐channel propagation measurements in the Ionian basin of the Mediterranean Sea have been analyzed to determine the attenuation coefficients in the frequency range of 50–3200 Hz. Concurrent measurements of sound‐speed, temperature, salinity, and pH show a strong sound channel having a broad minimum below 100 m and highly uniform properties over the 600‐km path. Explosive sources were detonated near the channel axis. The results obtained from hydrophones located near the axis are compared with predictions of the temperature/pH‐dependent relaxation‐absorption model (components: MgSO4, B(OH)3>, and MgCO3). Scattering loss is found to be minimal and agreement is good. Attenuation coefficients are compared with earlier values from the Ligurian Sea over a single refraction path approximately 35 km long.

It is well known that characterizing the noise field as isotropic may be an unrealistic assumption, particularly when estimating the performance of narrow beam arrays. In this paper, we investigate the effect of an anisotropic noise field on signal‐to‐noise ratio (SNR) as a function of range. A theoretical model that simulates the vertical directionality of wind‐generated ambient noise in the ocean has been developed and is demonstrated for frequencies between 1 and 24 kHz. Noise received by a Chebyshevweighted planar array is evaluated and displayed for a variety of measurement systems and environmental parameters, including beam elevation angle, array depth, water depth, bottom porosity, sound speed profile, and frequency. Additionally, the model demonstrates the sensitivity of the received noise to multipath energy. Using the noise model results, we then predict SNR versus range for both isotropic and anisotropic noise fields, with higher levels consistently seen in the anisotropic field. A comparison ...

Owing to the pressure dependence of the temperature of maximum density in fresh water, the sound channel in Lake Superior is extremely weak. The fast field program (FFP) is applied to determine if the bottom loss is significant for sound channel propagation in the kilohertz region. The values of attenuation are computed for the frequency range 500–10 000 Hz and compared to values obtained from the Lake Tanganyika experiment.

One of a series of sound propagation experiments recently conducted by the Naval Underwater Systems Center in the Gulf of Aaen took place across a basin with a sloping bottom. Three receiving hydrophones were located near one edge in 60 m of water. Results of this experiment in which bomb shots were used were compared with both a propagation model based on ray theory by J. P. Jones and P. W. Smith, Jr. (Bolt Beranek and Newman TM 84 of 7 July 1971) and a model based on normal mode theory. Since the received signal appeared as an impulse followed by a reverberant portion, each with different spatial properties, the two segments were treated separately.