Hudson Canyon Research Articles

A comparison of broadband and narrow‐band transmission loss in shallow water

Calibrated acoustic measurements were made at the Hudson Canyon Experiment site [L. Maiocco and W. Carey, J. Acoust. Soc. Am. Suppl. 1 86, S8 (1989)] during area characterization test II(ACT II) conducted in September 1993. This site was chosen because it is a surveyed area with known geophysical and geoacoustic properties. Transmission loss was measured along the previously run radials from bore hole 6010 using calibrated small broadband explosive charge sources and a narrow‐band continuous wave (cw) towed acoustic projector. These sources were, for the purposes considered here, co‐located in range because of low test ship speed (under 5 kn). The receiving sensor was a 25 element stationary vertical line array spanning the majority of the water column (73 m), with 19 elements extending off the bottom (60 m) and 6 elements along the bottom (30 m). Using the principle of reciprocity, the hydrophone elements at each of the source depths enabled comparisons of the transmission loss measurements for broadband and cw sources to be made over a range interval from 4–22 km at two frequencies (200 and 900 Hz). The comparisons were performed by means of linear regression of the data from the two sources. The results of the comparison will be presented and will include a detailed discussion of the precision of the results and of possible sources of experimental error. Finally, the cw transmission loss from ACT II will be compared with the previous measurements of Maiocco and Carey.

Multiple‐frequency robust adaptive matched‐field processing (MFP) in shallow water

Application of robust adaptive algorithms to multiple‐frequency MFP in shallow water has been studied and the minimum‐detectable‐level (MDL) has been examined via simulations. The multiple‐frequency MFP output is obtained by incoherently averaging the MFP outputs of the discrete narrow‐band‐components. Performance of three different averaging techniques (arithmetic‐mean, geometric‐mean, and harmonic‐mean) have been compared. The arithmetic‐mean averages the amplitudes, the geometric‐mean averages the dBs, and the harmonic‐mean averages the reciprocals of the amplitudes. These algorithms have been applied to shallow water data from the Hudson Canyon experiment and have demonstrated excellent source localization results. simulations show a 10‐ to 20‐dB improvement in the MDL against the simulated near or distant surface noise.

MODEL-BASED SIGNAL ENHANCEMENT FOR THE HUDSON CANYON EXPERIMENT

Model-based signal processing is a well-defined methodology enabling the inclusion of environmental (propagation) models, measurement (sensor arrays) models, and noise (shipping, measurement) models into a sophisticated processing algorithm. Here we investigate the design of a space-varying, nonstationary, model-based processor (MBP) for the Hudson Canyon experiment. In this shallow water application, a state space representation of the normal mode propagation model is used. The processor is designed such that it allows in situ recursive estimation of the both the pressure field and modal functions. It is shown that the MBP can be effectively utilized to "validate" the performance of the model on noisy ocean acoustic data. In fact, a set of processors is designed, one for each source range, and the results are reasonable, implying that the propagation model with measured parameters adequately represents the data.

Synthetic aperture experiment in the Hudson Canyon area

In 1988, narrow-band sound propagation measurements were performed in 73 m of water on the New Jersey Continental Shelf over a frequency range from 50 to 600 Hz. A mid-depth source was towed past a 24-element hydrophone array out to ranges of approximately 4 km. Precision time bases controlled the source and were recorded on one of the data channels. A Del Norte radar system measured the range to within several meters accuracy. Plots of the residual phase of the sound pressure as a function of range demonstrate the high quality of the data. A synthetic aperture technique was used to produce horizontal wave-number spectra at each of the hydrophone depths. Plots of the magnitude of the horizontal wave number as a function of depth and wave number clearly demonstrate the modal nature of the propagation. The sound speed in the water column was calculated from CTDs taken during the experiment, and the second speed in the bottom was derived using an acoustical inversion technique. The measured acoustical data along with the measured and derived environmental information provide a unique data set for testing processing and modeling algorithms. [Work supported by ONR.]

Source tracking in the Hudson Canyon experiment

In this shallow-water experiment a source was towed at midchannel depth to a range of 6 km from a vertical line array. The source simultaneously hummed four tones with pitches from 50 to 600 Hz. Using the matched-field approach, an acoustic model was used to scan a beam through range and depth in the channel. It is shown that the source can be tracked continuously by combining the data from all four frequencies.

Model-based processor design for a shallow water ocean acoustic experiment

Model-based signal processing is a well-defined methodology enabling the inclusion of environmental (propagation) models, measurement (sensor arrays) models, and noise (shipping, measurement) models into a sophisticated processing algorithm. Depending on the class of model developed from the mathematical representation of the physical phenomenology, various processors can evolve. Here the design of a space-varying, nonstationary, model-based processor (MBP) is investigated and applied to the data from a well-controlled shallow water experiment performed at Hudson Canyon. This particular experiment is very attractive for the inaugural application of the MBP because it was performed in shallow water at low frequency requiring a small number of modes. In essence, the Hudson Canyon represents a well-known ocean environment, making it ideal for this investigation. In this shallow water application, a state-space representation of the normal-mode propagation model is used. The processor is designed such that it allows in situ recursive estimation of both the pressure-field and modal functions. It is shown that the MBP can be effectively utilized to ‘‘validate’’ the performance of the model on noisy ocean acoustic data. In fact, a set of processors is designed, one for each source range and the results are quite good—implying that the propagation model with measured parameters adequately represents the data.

Mixing of shelf, slope and Gulf Stream water over the continental slope of the Middle Atlantic Bight

A combination of water masses was observed over the continental slope between Cape Hatteras and Hudson Canyon following the passage of a Gulf Stream meander. It was largely made up of fluid discharged from the Gulf Stream in the wake of the meander and also contained intrusions of Middle Atlantic Bight (MAB) shelf water and slope water. Its most prominent dynamical feature was an anticyclonic eddy that had an initial diameter of about 100 km but contained relatively weak currents, not greater than 30 cm s −1 in magnitude. Discharged Gulf Stream water made up the eddy's core, and a band of entrained shelf water circulated at its margin. This band was connected with another entrained shelf water band that flowed along the northern margin of the Gulf Stream. Transport of the shelf water at the edge of the eddy was estimated to be 1.3 × 10 5 m 3 s −1, which is comparable with the estimated alongshore transport of water over the MAB shelf. Vigorous mixing of both shelf water bands with surrounding discharged Gulf Stream water is indicated by the large differences of the latter's T/S properties from those found within the Gulf Stream. These differences were used to estimate salt fluxes into the shelf water band at the edge of the anticyclonic eddy, with results in the range of 1.9–6.6 × 10 −6 g salt cm −2 s −2. Density ratios and Richardson numbers determined from hydrographic data suggest that these salt fluxes were primarily the result of mixing by double diffusive processes, and that shear-induced turbulence combined with double diffusion to effect vertical mixing at the margins of the shelf water band next to the Gulf Stream.

Petroleum Exploration and the Atlantic OCS

The largest Atlantic outer continental shelf (OCS) lease sale was the first one, Sale 40 in 1976. Ninety-three Baltimore Canyon Trough petroleum leases were issued, and industry's winning bids total $1.1 billion. The highest bonus bids were for leases overlying the Schlee Dome, then called Great Stone Dome, a large structure with a very large fetch area. By 1981, seven dry wells on the dome moderated this initial flush of optimism. However, subeconomic quantities of gas and light oil were discovered on the nearby Hudson Canyon Block 598-642 structure. Now after 9 lease sales, 410 lease awards, and 46 exploration wells, United States Atlantic petroleum exploration activity is in a hiatus. Fifty-three leases remain active under suspensions of operation. Twenty-one lease blocks, about 50 mi offshore from Cape Hatteras, have been combined as the Manteo Exploration Unit. Mobil and partners submitted an exploration plant for the unit in 1989. The Atlantic OCS has petroleum potential, especially for gas. With only 46 exploration wells, entire basins and plays remain untested. During the present exploration inactivity, some petroleum evaluation of the Atlantic OCS continues by the Minerals Management Service and others. Similarities and differences are being documented between United States basins andmore » the Canadian Scotian Basin, which contains oil and gas in commercial quantities. Other initiatives include geochemical, thermal history, seismic stratigraphic, and petroleum system modeling studies. The gas-prone Atlantic OCS eventually may make an energy contribution, especially to nearby East Coast markets.« less

Monitoring of the Fate, Transport, and Effects of Sewage Sludge Disposed at the 106-Mile Deepwater Municipal Sludge Dump Site

Abstract Since 1985, the US Environmental Protection Agency (EPA) has conducted monitoring studies to determine the transport, fate, and effects of sewage sludge dumped at the 106-Mile Deepwater Municipal Sludge Dump Site (106-Mile Site). This paper describes EPA's 106-Mile Site monitoring activities and the results from six oceanographic surveys of the Site. Surveys have been conducted to track sewage sludge plumes and monitor dispersion and settling characteristics; study plume toxicity and contaminant levels; search for sludge and sludge constituents in surface waters in the area of the Site up 74 km (40 nautical miles) away; maintain a stationary real-time current meter near the Site; deploy and retrieve approximately 50 sediment traps and 17 current meters on 10 arrays from Hudson Canyon south to Delaware Canyon, at depths ranging from 1500 to 2800 m; and deploy satellite-tracked drifters. Results of studies completed to date have provided field data on the chemical and physical behaviour of sludge d...

Water discharged from the Gulf Stream north of Cape Hatteras

Satellite radiometer‐derived sea surface temperature images together with moored instrument and hydrographic survey data indicate that water ejected from the Gulf Stream often occupies the upper 200 m of the water column over the continental slope between Cape Hatteras and Hudson Canyon. At times this water resembles the energetic Gulf Stream frontal eddies commonly seen to the south of Cape Hatteras. However, many of the observed parcels of this water differ markedly from frontal eddies and Gulf Stream warm‐core rings and so appear to form a class of discharged Gulf Stream water distinct from any previously reported. These parcels generally cover a broad area (compared with frontal eddies), contain relatively weak currents (generally < 40 cm s−1 at 100 m), and are remarkably long‐lived (sometimes distinguishable for more than 2 months). Salinity anomaly distributions indicate that despite their persistence, these water masses contain intrusions of and mix with surrounding lower‐salinity water, particularly in the upper 50 m. Continuity of various tracers along σθ surfaces reveals that the discharged Gulf Stream water observed near the sea surface originated within the nutrient‐bearing stratum of the Gulf Stream, having upwelled hundreds of meters along density surfaces. As a result, it tends to enhance nutrient concentrations over the continental slope, to a significant extent at 100 m depth but only marginally at 50 m depth. By contrast, this water does not carry unusually large kinetic energy densities into the slope region, but apparently gives up a good deal of kinetic energy, on the order of 104 cm2 s−2 per unit mass, while leaving the Gulf Stream.

Predictability of acoustic intensity in shallow water at low frequencies using parabolic approximations

The accuracy of predictions for acoustic intensity in shallow water from parabolic approximation models are carefully examined. First, accuracy comparisons are made between analytic solutions to the Helmholtz equation and parabolic equations for propagation in a Pekeris waveguide. The parameter dependencies of the interference patterns for frequencies below 100 Hz are described. Calculations from the IFD implementation of parabolic equations are also compared with the analytic solutions. Agreement between the analytical and numerical results is very good, providing that appropriate initial conditions are employed. Sensitivities to input parameters are examined by observing the variations in the predicted fields and intensities as the parameters are changed. The parameters include those specifying the depth and range properties of the geoacoustic bottom, sound-speed profile, and spatial features of the waveguide. Magnitudes of parameter variations are selected consistent with a recent Hudson Canyon experiment. Comparisons of the experimental results with model predictions are discussed. One conclusion is that input parameter uncertainties, even though relatively small for this experiment, can in some cases significantly affect the predictions. [Work supported by ONR.]

Metals and organic contaminants in northwest Atlantic deep-sea tilefish tissues

Data are presented on trace metal and organic hydrocarbon (PCBs, PAHs, and chlorinated pesticides) concentrations in muscle, liver, gonad, and other tissues from two small 1981–1982 collections of tilefish ( Lopholatilus chamaeleonticeps): from near Lydonia Canyon (on Georges Bank) and below Hudson Canyon (off New Jersey). Significant concentrations of trace metals and chlorinated hydrocarbons (PCBs and pesticides) were detected in the tissues, but these were generally within the ranges reported for other deep-sea and continental shelf fish in the Middle Atlantic Bight. Higher chlorinated hydrocarbon concentrations were detected in fish from the Lydonia Canyon area, compared with concentrations detected in the fish from below the Hudson Canyon. The difference appears to be statistically significant for most compounds and no satisfactory explanation can be suggested.

Measurement and modeling of acoustic waves in the 50- to 600-Hz range at the Hudson Canyon site

Measurements of the shallow-water propagation of acoustic waves on the New Jersey continental shelf were conducted by the Naval Underwater Systems Center, September 1988. On the same site, accurate measurements of geoacoustic properties were obtained by Yamamoto et al. (1988) using the bottom shear modulus profiler (BSMP) method invented by T. Yamamoto. The bottom properties extracted by the BSMP inversion method were introduced to SAFARI computer code for modeling of normal-mode acoustic propagation by the author. Favorable agreement has been found between model calculations and results of the acoustic propagation experiment conducted by NUSC. Comparison has revealed findings about the contributions of the compressional and shear wave characteristics of the seabed to the acoustic waveguide. The compressional wave profile has been found to contribute to the characteristics of the acoustic waveguide modal structure, while the shear wave profile has been found to contribute only to the attenuation of the acoustic wave propagation. [This work sponsored by the Office of Naval Research.]

Uptake and effects of copper and cadmium in the gonad of the scallop Placopecten magellanicus: concurrent metal exposure

Sea scallops Placopecten magellanicus in early gametogenesis from the southern shelf of Hudson Canyon, New Jersey, USA, were exposed in late winter 1984 to sublethal levels of Cu and Cd in a flowing seawater system at the NMFS Milford Laboratory. Exposure was to copper (10 and 20 μgl-1: low-Cu and high-Cu groups) or to a combination of copper and equimolar cadmium (10μg Cu+17.7 μg Cdl-1: low-Cu/Cd group) for eight weeks, with sampling at 2-wk intervals. Copper had a strongly inhibitory effect on gamete production and maturation, which in some respects was partially moderated in the presence of cadmium in the female gonad only. Total gamete weight per scallop doubled in control individuals but dropped by 60% in both high-metal exposure groups over the 8-wk exposure period, with a smaller, temporary decrease in the low-Cu group. Cadmium did not add to the inhibition by copper of gamete development in the low-Cu/Cd group, but there was no partial recovery at 8 wk, as was seen in female scallops exposed to low-Cu alone. Gonadal RNA, higher in the females, decreased proportionately more in that sex than in the males of the metal exposure groups. Conversely, DNA levels were higher in the male than in the female gonad, and decreased sharply in all metal-exposed males. Gonadal protein concentration also dropped in all metal-exposed scallops with time and degree of metal exposure. Copper uptake in the gonad increased with time and metal exposure concentration, and cadmium increased in the low-Cu/Cd group with time. In contrast, manganese decreased significantly in the gonads of Cu-exposed scallops, especially in the high-Cu group. In the low-Cu/Cd group, manganese concentrations stabilized after an initial sharp drop at 2 wk in the low-Cu/Cd group, then rose at 6 and 8 wk. We ascribe this phenomenon to the induction by cadmium of metal-binding proteins (Fowler and Megginson 1986). In both sexes, gonadal magnesium concentrations did not change with increasing tissue burdens of copper and cadmium, but instead rose initially in proportion to the degree of metal exposure, homeostasis being maintained thereafter.

Sources and Distribution of Upper Pleistocene Sand, Eastern United States Atlantic Shelf: ABSTRACT

A 2-yr study of the sources and distribution of upper Pleistocene and Holocene sand on the eastern United States shelf between the Bay of Fundy and Cape Hatteras reveals that 3 sand types are found on this shelf: (1) glacially transported, very angular sands, (2) fluvially transported, well-rounded sands derived from unlithified coastal plain deposits, and (3) fluvially transported, moderately angular sands derived from lithified sedimentary and crystalline rocks of the Appalachian and New England areas. For the most part, the distribution of these sand types reflects the late Pleistocene paleogeography of this shelf. Glacial sands are found in the areas of upper Pleistocene till, moraine, and outwash-plain deposits east and northeast of the Hudson Canyon; the 2 fluvial s nds are found in coast-normal stripes that correspond to the ancestral paths of the many rivers that traversed this shelf during the late Pleistocene. The preservation of relict paleogeographic patterns of these sorts are an indication of diffusive transport of sand through most of this shelf. The exceptions to this are found in the shallow waters of Nantucket Shoals and Georges Bank, where glacial sands are presently being advected to the southwest by the strong tidal currents that prevail. End_of_Article - Last_Page 278------------

The Role of Erosion by Fish in Shaping Topography Around Hudson Submarine Canyon

ABSTRACT An 800-km2 area of rough topography around the head of Hudson Canyon off the eastern United States is attributed to erosion by tilefish (Lopholatilus chamaeleonticeps) and associated species of crustaceans. The rough topography has a relief of 1-10 m, occurs in water depths of 120-500 m, and has been cut into a semilithified, silty clay substrate since the onset of the Holocene transgression. Commercial fishing activity indicates that a large population of tilefish, which dig burrows in the sea floor, occupy the area of the rough topography. Average tilefish burrows are 1.6 m in diameter and 1.7 m in depth. They have a clustered, not uniform, distribution, and their average density is 2,500 per km2. The close match of areas of rough topography and high til fish populations, the active burrowing of the sea floor, and the clustered distribution of the burrows suggest that the hummocky topography in this area may be the result of continuous erosion by tilefish and associated crustaceans during the Holocene. An erosion rate of 13 cm per 1,000 years is necessary to create this topography during the past 13,000 years--and 18 cm per 1,000 years if(as is more likely based on the depths at which tilefish presently are found) the erosion started 9,000 years ago.

Sedimentary processes on the Atlantic Continental Slope of the United States

Until recently, the sedimentary processes on the United States Atlantic Continental Slope were inferred mainly from descriptive studies based on the bathymetry and on widely spaced grab samples, bottom photographs, and seismic-reflection profiles. Over the past 6 years, however, much additional information has been collected on the bottom morphology, characteristics of shallow-subbottom strata, velocity of bottom currents, and transport of suspended and bottom sediments. A review of these new data provides a much clearer understanding of the kinds and relative importance of gravitational and hydrodynamic processes that affect the surface sediments. On the rugged slope between Georges Bank and Cape Lookout, N.C., these processes include: (1) small scale mass wasting within submarine canyons and peripheral gullies; (2) density flows within some submarine valleys; (3) sand spillover near the shelf break; (4) sediment creep on the upper slope; and (5) hemipelagic sedimentation on the middle and lower slope. The area between Georges Bank and Hudson Canyon is further distinguished by the relative abundance of large-scale slump scars and deposits on the open slope, the presence of ice-rafted debris, and the transport of sand within the heads of some submarine canyons. Between Cape Lookout and southern Florida, the slope divides into two physiographic units, and the topography is smooth and featureless. On the Florida—Hatteras Slope, offshelf sand spillover and sediment winnowing, related to Gulf Stream flow and possibly to storm-driven currents, are the major processes, whereas hemipelagic sedimentation is dominant over the offshore slope along the seaward edge of the Blake Plateau north of the Blake Spur. Slumping generally is absent south of Cape Lookout, although one large slump scarp (related to uplift over salt diapirs) has been identified east of Cape Romain. Future studies concerning sedimentary processes on the Atlantic slope need to resolve: (1) the ages and mechanisms of mass wasting; (2) the accumulation rates and thicknesses of hemipelagic sediments; and (3) the causes and variability of offshelf sand spillover, sediment winnowing, and canyon transport.

The fish fauna of the Galapagos thermal vent region

About 20 species of fishes live in the vacinity of the Galapagos Rift thermal springs at depths of about 2,400 m at about 0°48′N, 86°08′W. Records are based on sightings from a manned submersible, time-lapse photographs near a baited trap, and actual captures. The five or six most abundant species were Coryphaenoides bulbiceps, Bassozetus, one or two zoarcids, Antimora, and Spectrunculus. Only a single species, an undescribed bythitid, possibly in the genus Diplacanthopoma, lives in the vents. The thermal vent fish fauna is about as diverse as that of the flat bottom near Hudson Canyon at comparable depths. Relatively few fishes were attracted to bait. Endemism seems lower among the fishes than invertebrates. Species diversity and population density of fishes overall decrease closer to the vents. Opposite trends appear to characterize invertebrates.

The Middle Atlantic Bight Cold Pool: Evolution of the Temperature Structure During Summer 1979

Temperature data spanning the entire Middle Atlantic Bight (MAB) during 1979 are used to study the structure and evolution of the cold pool. The Nantucket Shoals and New England Shelf appear to be the source of the coldest water found in the MAB in late winter. During the spring and summer, water within the cold pool in the New York Bight north of Hudson Canyon remains colder than any shelf water either to the northeast or southwest. Thus the coldest cold-pool water persists there as a remnant of winter-cooled water rather than being replenished by a colder upstream source, and south of Hudson Canyon, cold-pool temperatures decrease in June and July as colder water from upstream is advected southwestward along the coast. Both temperature data and direct current measurements suggest that the mean alongshore current has a minimum between Nantucket Shoals and Hudson Canyon. The alongshore variation of shelf topography appears to be responsible for the spatial variation in both the alongshelf drift speed and the thermal structure of the cold pool.

Local and distal origin of sands in the Hatteras Abyssal Plain

Quartz-grain shape analysis and SEM imagery of Pleistocene sediment in the northern and central Hatteras Abyssal Plain indicates two sources for very fine sand in the plain. The first source contributed a diagnostic suite of relatively well-rounded grains with many silica-solution and precipitation features on the surface, indicating their derivation from pedologic environments such as the soils of the southeastern U.S. Piedmont and coastal plain. The second source was apparently a glacially affected terrain, as evidenced by the plethora of irregularly shaped quartz grains with conchoidal fracture features. One of the known sources for sediment in the Hatteras Abyssal Plain is the Hatteras Canyon System (Elmore et al., 1979), which drains part of the continental margin of the southeastern U.S. Not surprisingly, an analysis of 29 samples in the Black Shell Turbidite (Elmore et al., 1979), a deposit of known Hatteras Canyon origin, showed that their sands were composed predominantly (73–97%) of the well-rounded grains of the first source and none of the grains of the second source. Thus, the presence of high proportions of well-rounded very fine sand grains was considered to be indicative of sediment transported into the basin via the Hatteras Canyon System. Sixteen other sand layers, all confined to an area within 150 km of the mouth of the Hatteras Canyon, also contained high proportions of this sand type, and hence were attributed to Hatteras Canyon turbidity currents. Eastward and southward of this area, however, very fine sand grains produced in glacially affected areas predominated in all the sand layers sampled (except the Black Shell Turbidite). It is apparent from this study, therefore, that the sand supplied to the northern and central Hatteras Abyssal Plain was brought into the basin not only via the Hatteras Canyon System, as has been shown by Elmore et al. (1979), but from other, more northern canyon systems as well. Likely candidates for these northern transport paths include the Wilmington Canyon System, which was supplied with sediment by the ancestral Hudson River in the Pleistocene (Kelling et al., 1975), the Hudson Canyon System (Horn et al., 1971), and possibly canyon systems further north (Pratt, 1968).