Downslope Conversion Research Articles

Contributions of gravity waves in the ocean to T-phase excitation by earthquakes.

The generation of T waves in a deep ocean by an earthquake in its epicentral region is often observed, but the mechanism of the excitation of the acoustic waves travelling horizontally with the speed of sound remains controversial. Here, the hypothesis is investigated that the abyssal T waves are generated by the scattering of ballistic sound waves by surface and internal gravity waves in the ocean. Volume and surface scattering are studied theoretically in the small perturbation approximation. In the 3-50 Hz typical frequency range of the observed T waves, the linear internal waves are found to lack the necessary horizontal spatial scales to meet the Bragg scattering condition and contribute appreciably to the T-wave excitation. In contrast, the ocean surface roughness has the necessary spatial scales at typical sea states and wind speeds. The efficiency of the acoustic normal modes' excitation at surface scattering of the ballistic body waves by wind seas and sea swell is quantified and found to be comparable to that of the established mechanism of the T-wave generation at downslope conversion at the seamounts. The surface scattering mechanism is consistent with key observational features of the abyssal T waves, including their ubiquity, low-frequency cutoff, presence on seafloor sensors, and weak dependence on the earthquake focus depth.

Comparison of scattering mechanisms for excitation of axial modes by surface sources

The vertical spectrum of low-frequency ambient noise in deep water contains significant contributions from distant shipping. Distant noise is concentrated at angles around broadside that are associated with the low order modes. Near-surface sources cannot excite the low modes directly. Dashen and Munk suggest three mechanisms for transferring energy from high modes to low modes: internal wave (IW) scattering, downslope conversion, and excitation at high latitudes where the sound channel intersects the surface. Dashen and Munk conclude that slope conversion is the most likely due to the weakness of IW scattering and the lack of shipping traffic at high latitudes. Commenting on their IW analysis, Dashen and Munk note that they are not totally convinced by their theoretical results and admit that IW scattering may have a significant effect. This talk compares the IW scattering and downslope conversion mechanisms for transferring surface source energy to the low modes. IW scattering is modeled using transport theory [Colosi and Morozov, JASA 2009] and compared to parabolic equation and coupled mode simulations. Results show that IW scattering from a large number of ships can produce vertical spectra comparable to those obtained through downslope conversion. [Work supported by ONR.]

The effect of internal waves on the ambient noise vertical directionality in deep ocean

Vertical array measurements of low-frequency ambient noise in the mid-latitudes indicate that distant shipping noise has a flat angular distribution concentrated around broadside, e.g., [Wales and Diachok, JASA, 1981]. Ships cannot directly excite the lowest modes (associated with broadside angles) since those modes are trapped near the sound channel axis. Dashen and Munk [JASA, 1984] considered three mechanisms that could transfer shipping noise into the low modes: (1) downslope conversion on the continental slope, (2) volume scattering due to internal waves, and (3) direct excitation of the low modes at high latitudes where the sound channel intersects the surface. Dashen and Munk conclude that the first mechanism is the most likely since the second requires impractically large propagation ranges and the third requires unrealistic shipping densities at high latitudes. A limitation of Dashen and Munk's internal wave analysis is that it does not consider the effect of bottom loss. This talk adapts the transport theory approach of Colosi and Morozov [JASA, 2009] to study the effect of internal waves on the vertical directionality of distant shipping noise in the presence of seafloor attenuation. Initial results indicate that internal waves may have a significant effect at shorter ranges than previously thought.

Effects of seismic source and environment parameters on elastic bottom parabolic equation solutions

Recently, two-types of elastic self-starters have been incorporated into parabolic equation solutions for range-dependent elastic bottom underwater acoustic problems. These source fields generate parabolic equation solutions that can be used to study development of oceanic T-phases via the process of downslope conversion. More general range-dependence has also been shown to scatter elastic wave energy into water column acoustic modes which then propagate as T-phases. In certain circumstances, sources in the elastic bottom can also cause interface waves at the ocean bottom that contribute to the ocean acoustic field. Both types of waves can propagate long distances and could be source mechanisms for unexplained acoustic signals recorded near the sea floor and below the ray-theoretic turning point. Parabolic equation solutions will be used to study effects of parameters such as frequency, source location, and source type on T-phase and interface wave generation and propagation.[Work supported by ONR.]

Horizontal noise directionality in deep water measurements and modeling.

Low‐frequency, less than 100 Hz, sound is dominated by surface shipping, particularly in the northern hemisphere. With low‐volume attenuation sound can propagate significant distances. For periods of time not dominated by a particular nearby interferer, the low‐frequency ambient noise field is dominated by distant shipping and is expected to be anisotropic due to various bathymetric effects such as downslope conversion and seamount or ocean ridge blockage. In this paper, the authors present horizontal array measurements from a 100 m towed line array (towed at 100 m) in very deep water in the Philippine Sea. The northern Philippine Sea has several ridges and seamounts and significant bathymetric interaction for surface ships in the main Singapore Shanghai shipping lane. A rapid range bearing dependent N×2D parabolic equation approach will be described and used to evaluate the bathymetric effects on the horizontal directionality of low‐frequency noise. Comparisons of measurements and models will be done using both Monte‐Carlo realizations of surface ship positions (from the HITS database) as well as measured AIS from research vessels in the vicinity.

Fred Fisher and research with acoustic vector sensors; Marine Physical Laboratory’s vertical array of directional acoustic receivers and ocean noise

Fred Fisher had boundless enthusiasm for all topics acoustic. A chance encounter with him in the hallway usually led to a half-hour discussion of the latest research efforts at the lab and recent results he found exciting. In the 1980s, Fred became interested in the problem of identifying the physical phenomena forming the pedestal about the horizontal in vertical directionality measurements of the deep ocean’s low-frequency noise field. Two competing mechanisms had been proposed: downslope conversion of coastal shipping and noise from high latitude winds coupling into the deep sound channel due to the shoaling of the sound channel axis. The relative contributions of these two mechanisms possibly could be separated if the azimuthal ambiguity of a vertical line array of hydrophones somehow could be broken. Therefore, Fred proposed to build a vertical array of ‘‘DIFAR’’ sensors, which led to the design and construction of the Marine Physical Lab’s Vertical ‘‘DIFAR’’ Array. This talk will reminisce a bit about Fred as well as present some results from an ambient noise experiment conducted in 1992 on the continental shelf using the Vertical DIFAR Array co-deployed with MPL’s freely drifting vector sensors, the Swallow floats. [Work supported by ONR and ONT.]

Downslope propagation of adiabatic and stepwise coupled modes

Downslope propagation, from a shallow source, redirects steeply propagating modes toward the horizontal. Slope conversion is of particular interest for noise sources near the shelf break. Noise modeling procedures may rely on adiabatic modes for efficiency, but their ability to accurately handle downslope conversion is not well tested. The adiabatic approximation is derived from continuously coupled modes by assuming the coupling matrices are the identity matrix and by using an asymptotic approximation to solve the uncoupled ordinary differential equations for the modal coefficients. A simpler interpretation of the twofold adiabatic approximation is obtained in the context of stepwise coupled modes when the continuous differential equations are replaced by a single matrix recursion. The stepwise interpretation of the adiabatic approximation is used to evaluate its validity, for a shelf-break environment, in a side-by-side comparison with stepwise coupled modes. The adiabatic approximation partially accounts for downslope conversion, but vertical directivity, computed in with adiabatic modes and stepwise coupled modes, is not comparable. Parabolic equation calculations, which corroborate the stepwise coupled mode results, provide a practical alternative to adiabatic modes for ambient noise modeling in shelf-break environments.

Characteristics of T phases from earthquake sources, or how earthquakes ‘‘talk’’ into the ocean

We review the mechanisms by which earthquake sources within the solid Earth generate acoustic energy into the SOFAR channel. We examine individual cases involving both the relatively inefficient mechanism of ‘‘downslope conversion,’’ leading to T phases with low amplitudes and long durations, and efficient conversion at steep interfaces allowing direct penetration of the SOFAR, larger amplitudes, and shorter durations. We derive several ways to quantify T waves generated by an earthquake source. To avoid the saturation of their amplitude as source size grows, we use the concept of a T-phase energy flux (TPEF), which mirrors the radiated energy measured on conventional seismic body waves. TPEF is scaled to the seismic moment of the earthquake to define a parameter, gamma, characterizing the earthquake’s efficiency for T-phase generation. In another approach, we compare the amplitude of T-phase envelopes to the duration of their signal. The resulting discriminant has been proposed to identify earthquakes from explosions. We show that it varies for several categories of nonstandard earthquakes, notably those occurring in volcanic islands and events with slow rupture giving rise to large tsunamis. Finally, we discuss a few examples of T waves from underwater landslides.

Contribution of bottom-refracted sound to oceanic sound propagation

In the deep ocean, sound from an underwater source is channeled in the SOFAR channel to form a long series of convergence zones. These may be influenced by bottom reflections when the propagation occurs down a sloping ocean floor. Experiments [W. M. Carey, J. Acoust. Soc. Am. 79, 49 (1986)] show important residual intensities (∼10 dB excess) between convergence zones whose origin could not be conclusively explained by downslope conversion, but could have been due to ‘‘refraction’’ (bottom penetration). Earlier pulse return experiments by Christensen etal. [J. Acoust. Soc. Am. 57, 421 (1975)] conclusively established the presence of bottom penetration effects in the presence of upward-refracting sound-speed gradients in ocean-floor sediments from a flat, deep (4000-m) ocean floor. In the present model calculation, carried out for the Christiansen deep-water case by using normal-mode and fast-field models, the influence of a shear-supporting, upward-refracting ocean floor on oceanic sound propagation, including the long-range presence of inter-convergence zone residual sound fields caused by bottom penetration is demonstrated. [Work supported by the ONT.]

Downslope conversion in oceanic sound propagation: CENTRO/ANTS code interpretation

The 1989 experiment by the Scripps Marine Physical Laboratory, studying sound propagation down the continental slope off Pt. Arguello and Pt. Conception, CA, has furnished evidence for the downslope conversion effect in which sound propagating down from a source over the continental shelf, by its interaction with the sloping ocean floor, gets trapped into the sound channel. As a consequence, the horizontal arrival structure observed at a distant receiver over deep water (the Scripps FLIP array, in this case) shows a flat ‘‘pedestal’’ shape centered at ±20° about the horizontal, while with a source also over deep water, a central minimum would be apparent due to Snell’s law. Our calculations, using the CENTRO/ANTS normal mode code written by J. I. Arvelo and developed further at Catholic University, bears out the pedestal shape of the arrival structure. In these calculations, downslope propagation is treated adiabatically, and the bottom structure (sediment with underlying rock) is correctly taken into account. [Work supported by the ONR.]

Mid‐ocean basin manifestation of continental margin downslope converted energy

In July 1989, the Marine Physical Laboratory carried out a downslope conversion experiment in the region of the continental margin NW of Pt. Conception, CA. These carefully controlled and well‐documented measurements of downslope signal propagation were made to study the physics of downslope propagation, which has been suggested as one mechanism by which acoustic energy from surface sources such as storms and shipping traffic gets coupled into the sound channel [e.g., R. A. Wagstaff, J. Acoust. Soc. Am. 69, 1009–1004 (1981)]. At long distance, this downslope converted energy is seen as a large contribution to the ambient noise level arriving near the horizontal. This paper will discuss the downslope conversion experiment and will compare the observed vertical arrival structure from a source in deep water and a source traveling across the continental margin. [Work supported by ONR.]

Simulation of array processing in three-dimensional noise fields: Suggested experiments

The growing capability to model three-dimensional noise fields in complex ocean environments taken together with concurrent recent developments in matched-field processing (MFP) suggests noise experiments that will shed light onto the physical processes generating and propagating the noise fields. These experiments should especially yield information concerning the noise field as it impacts on signal and/or array processing. Some of the expected (modeled) characteristics of low-frequency ambient noise in a deep ocean basin with realistic complexity and its subsequent impact on MFP are reviewed. For example, the effects on MFP of down-slope conversion from discrete sources (surface ships) and distributed sources (storms), and the effectiveness of various rejection techniques are demonstrated. These results suggest experiments to investigate the most important processes.