ALBERT

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Physical Oceanography at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2019.

Description: Each year, surface ocean ecosystems export sinking particles containing gigatons of carbon into the ocean’s interior. This particle flux connects the entire ocean microbiome and constitutes a fundamental aspect of marine microbial ecology and biogeochemical cycles. Particle flux is also variable and intricately complex, impeding its mechanistic or quantitative description. In this thesis we pair compilations of available data with novel mathematical models to explore the relationships between particle flux and other key variables – temperature, net primary production, and depth. Particular use is made of (probability) distributional descriptions of quantities that are known to vary appreciably. First, using established thermodynamic dependencies for primary production and respiration, a simple mechanistic model is developed relating export efficiency (i.e. the fraction of primary production that is exported out of the surface ocean via particle flux) to temperature. The model accounts for the observed variability in export efficiency due to temperature without idealizing out the remaining variability that evinces particle flux’s complexity. This model is then used to estimate the metabolically-driven change in average export efficiency over the era of long-term global sea surface temperature records, and it is shown that the underlying mechanism may help explain glacial-interglacial atmospheric carbon dioxide drawdown. The relationship between particle flux and net primary production is then explored. Given that these are inextricable but highly variable and measured on different effective scales, it is hypothesized that a quantitative relationship emerges between collections of the two measurements – i.e. that they can be related not measurement-by-measurement but rather via their probability distributions. It is shown that on large spatial or temporal scales both are consistent with lognormal distributions, as expected if each is considered as the collective result of many subprocesses. A relationship is then derived between the log-moments of their distributions and agreement is found between independent estimates of this relationship, suggesting that upper ocean particle flux is predictable from net primary production on large spatiotemporal scales. Finally, the attenuation of particle flux with depth is explored. It is shown that while several particle flux-versus-depth models capture observations equivalently, these carry very different implications mechanistically and for magnitudes of export out of the surface ocean. A model is then proposed for this relationship that accounts for measurements of both the flux profile and of the settling velocity distribution of particulate matter, and is thus more consistent with and constrained by empirical knowledge. Possible future applications of these models are discussed, as well as how they could be tested and/or constrained observationally.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Oceanography and Applied Ocean Science and Engineering at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2019.

Description: Submesoscale currents, with horizontal length scales of 1-20 km, are an important element of upper ocean dynamics. These currents play a crucial role in the horizontal and vertical redistribution of tracers, the cascade of tracer variance to smaller scales, and in linking the mesoscale circulation with the dissipative scales. This thesis investigates submesoscale flows and their properties using Lagrangian trajectories of observed and modeled drifters. We analyze statistics of observed drifter pairs to characterize turbulent dispersion at submeso-scales. Contrary to theoretical expectations, we find that nonlocal velocity gradients associated with mesoscale eddies dominate the separation of drifters even at the kilometer scale. At submeso-scales, we observe energetic motions, such as near-inertial oscillations, that contribute to the energy spectrum but are inefficient at dispersion. Using trajectories in a model of submesoscale turbulence, we find that, if drifters have a vertical separation, vertical shear dominates the dispersion and conceals horizontal dispersion regimes from drifter observations. Particularly in submesoscale flows, vertical shear is orders of magnitude larger than horizontal gradients in velocity. Since conventional drifters in the ocean are not affected by vertical shear, it is likely that drifter-derived diffusivity underestimates the diffusivity that a tracer would experience. Lastly, we test and apply cluster-based methods, using three or more drifters, to estimate the velocity gradient tensor. Since velocity gradients become large at submesoscales, the divergence, strain, and vorticity control the evolution and deformation of clusters of drifters. Observing the velocity gradients using drifters, enables us to further constrain the governing dynamics and decipher submesoscale motions from inertia-gravity waves. These insights provide a Lagrangian perspective on submesoscale flows that illuminates scales that are challenging to observe from other platforms. We reveal observational and theoretical challenges that need to be overcome in future investigations.

Description: My doctoral studies in the WHOI/MIT Joint Program were funded by the National Science Foundation (OCE-I434788) and the Office of Naval Research (N00014-13-1-0451, Grant N00014-16-1-2470).

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June, 1982

Description: Oceanic fluctuations are dependent on geographical location. Near intense currents, the eddy field is highly energetic and has broad meridional extent. It is likely that the energy arises from instabilities of the intense current. However, the meridional extent of the linearly most unstable modes of such intense jets is much narrower than the observed region of energetic fluctuations. It is proposed here that weaker instabilities, in the linear sense, which are very weakly trapped to the current, may be the dominant waves in the far field. As a preliminary problem, the (barotropic) instability of parallel shear flow on the beta plane is discussed. An infinite zonal flow with a continuous cross-stream velocity gradient is approximated with segments of uniform flow, joined together by segments of uniform potential vorticity. This simplification allows an exact dispersion relation to be found. There are two classes of linearly unstable solutions. One type is trapped to the source of energy and has large growth rates. The second type are weaker instabilities of the shear flow which excite Rossby waves in the far field: the influence of these weaker instabilities extends far beyond that of the most unstable waves. The central focus of the thesis i: the linear stability of thin, twolayer, zonal jets on the beta plane, with both horizontal and vertical shear. The method used for the parallel shear flow is extended to the two-layer flow. Each layer of the jet has uniform velocity in the center, bordered by shear zones with zero potential vorticity gradient. The velocity in each layer outside the jet is constant in latitude. Separate linearly unstable modes arise from horizontal and vertical shear. The energy source for the vertical shear modes is nearly all potential while the source for the horizontal shear modes is both kinetic and potential. The most unstable waves are tightly trapped to the jet, within two or three deformation radii for small but nonzero beta. Rossby waves and baroclinically unstable waves (in the presence of vertical shear) exist outside the jet because of a nonzero potential vorticity gradient there. Weakly growing jet instabilities can force these waves when their phase speeds and wavelengths match. In particular, westward jets and any jets with vertical shear exterior to the jet can radiate in this sense. The radiating modes influence a large region, their decay scales inversely proportional to the growth rate. Two types of radiating instability are found: (1) a subset of the main unstable modes near marginal stability and (2) modes which appear to be destabilized neutral modes. Westward jets have more vigorously unstable radiating modes. Applications of the model are made to the eddy field south of the Gulf Stream, using data from the POLYMODE settings along 55°W and farther into the gyre at MODE. The energy decay scale and the variation of vertical structure with latitude in different frequency bands can be roughly explained by the model. The lower frequency disturbances decay more slowly and become more surface intensified in the far field. These disturbances are identified with the weak, radiating instabilities of the model. The higher frequency disturbances are more trapped and retain their vertical structure as they decay, and are identified with the trapped, strongly unstable modes of the jet.

Description: This work was supported by a grant from the National Science Foundation, Office of Atmospheric Science.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2017

Description: The Gulf Stream and Deep Western Boundary Current (DWBC) shape the distribution of heat and carbon in the North Atlantic, with consequences for global climate. This thesis employs a combination of theory, observations and models to probe the dynamics of these two western boundary currents. First, to diagnose the dynamical balance of the Gulf Stream, a depth-averaged vorticity budget framework is developed. This framework is applied to observations and a state estimate in the subtropical North Atlantic. Budget terms indicate a primary balance of vorticity between wind stress forcing and dissipation, and that the Gulf Stream has a significant inertial component. The next chapter weighs in on an ongoing debate over how the deep ocean is filled with water from high latitude sources. Measurements of the DWBC at Line W, on the continental slope southeast of New England, reveal water mass changes that are consistent with changes in the Labrador Sea, one of the sources of deep water thousands of kilometers upstream. Coherent patterns of change are also found along the path of the DWBC. These changes are consistent with an advective-diffusive model, which is used to quantify transit time distributions between the Labrador Sea and Line W. Advection and stirring are both found to play leading order roles in the propagation of water mass anomalies in the DWBC. The final study brings the two currents together in a quasi-geostrophic process model, focusing on the interaction between the Gulf Stream’s northern recirculation gyre and the continental slope along which the DWBC travels. We demonstrate that the continental slope restricts the extent of the recirculation gyre and alters its forcing mechanisms. The recirculation gyre can also merge with the DWBC at depth, and its adjustment is associated with eddy fluxes that stir the DWBC with the interior. This thesis provides a quantitative description of the structure of the overturning circulation in the western North Atlantic, which is an important step towards understanding its role in the climate system.

Description: My research was funded by National Science Foundation grants OCE-0241354, OCE- 0726720 and OCE-1332667 as well as a graduate fellowship from the American Meteorological Society. Support for travel and educational supplies was also provided by the MIT Houghton Fund and the WHOI Academic Programs Office.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2017

Description: The daily heating of the ocean by the sun can create a stably stratified near-surface layer when the winds are slight and solar insolation is strong. This type of shallow stable layer is called a Diurnal Warm Layer (DWL). This thesis examines the physics and dynamics of DWLs from observations of the subtropical North Atlantic Ocean associated with the Salinity Processes in the Upper ocean Regional Study (SPURS-I). Momentum transferred from the atmosphere to the ocean through wind stress becomes trapped within the DWL, generating shear across the layer. During SPURS-I, strong diurnal shear across the DWL was coincident with enhanced turbulent kinetic energy (TKE) dissipation (𝜖, 𝜖 〉 10−5 W/kg) observed from glider microstructure profiles of the near-surface. However, a scale analysis demonstrated that surface forcing, including diurnal shear, could not be the sole mechanism for the enhanced TKE dissipation. High-frequency internal waves (𝜔 ≫ 𝑓) were observed in the upper ocean during the daytime within the DWL. Internal waves are able to transfer energy from the deep ocean into the DWL through the unstratified remnant mixed layer, which is the intervening layer between the DWL and seasonal thermocline. As the strength of the stratification of the DWL increases, so does the shear caused by the tunneling internal waves. The analysis demonstrates that internal waves can generate strong enough shear to cause a shear-induced instability, and are a plausible source of the observed enhanced TKE dissipation. Vertically-varying horizontal transport across the upper ocean occurs because a diurnal current exists within the DWL, but not in the unstratified remnant mixed layer below. Therefore, when a DWL is present, the water within DWL is horizontally transported a different distance than the water below. Coupled with nocturnal convection that mixes the DWL with the unstratified layer at night, this cycle is a mechanism for submesoscale (1-10 km) lateral diffusion across the upper ocean. Estimates of a horizontal diffusion coefficient are similar in magnitude to current estimates of submesoscale diffusion based on observations, and are likely an important source of horizontal diffusion in the upper ocean.

Description: Supported by the Department of Defense (DoD) through the National Defense Science & Engineering Graduate Fellowship (NDSEG) Program and the National Science Foundation under Grant No. OCE-1129646. The collection and analysis of data from the SPURS-I central mooring were supported under National Aeronautics and Space Administration (NASA) Grant No. NNX11AE84G and NNX14AH38G.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2015

Description: The spatial and temporal evolution of stratified shear instabilities is quantified in a highly stratified and energetic estuary. The measurements are made using high-resolution acoustic backscatter from an array composed of six calibrated broadband transducers connected to a six-channel high-frequency (120-600 kHz) broadband acoustic backscatter system. The array was mounted on the bottom of the estuary and looking upward. The spatial and temporal evolution of the waves is described in terms of their wavelength, amplitude and turbulent dissipation as a function of space and time. The observed waves reach an arrested growth stage nearly 10 times faster than laboratory and numerical experiments performed at much lower Reynolds number. High turbulent dissipation rates are observed within the braid regions of the waves, consistent with the rapid transition to arrested growth. Further, it appears that the waves do not undergo periodic doubling and do not collapse once their maximum amplitude is reached. Under some conditions long internal waves may provide the perturbation that decreases the gradient Richardson number so as to initiate shear instability. The initial Richardson number for the observed instabilities is likely between 0.1 and 0.2 based on the slope and growth rate of the shear instabilities.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2014

Description: Motivated by inspection of complex underwater environments, we have developed a system for multi-sensor SLAM utilizing both structured and unstructured environmental features. We present a system for deriving planar constraints from sonar data, and jointly optimizing the vehicle and plane positions as nodes in a factor graph. We also present a system for outlier rejection and smoothing of 3D sonar data, and for generating loop closure constraints based on the alignment of smoothed submaps. Our factor graph SLAM backend combines loop closure constraints from sonar data with detections of visual fiducial markers from camera imagery, and produces an online estimate of the full vehicle trajectory and landmark positions. We evaluate our technique on an inspection of a decomissioned aircraft carrier, as well as synthetic data and controlled indoor experiments, demonstrating improved trajectory estimates and reduced reprojection error in the final 3D map.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2014

Description: This thesis explores the buoyancy-driven circulation in the Red Sea, using a combination of observations, as well as numerical modeling and analytical method. The first part of the thesis investigates the formation mechanism and spreading of Red Sea Overflow Water (RSOW) in the Red Sea. The preconditions required for open-ocean convection, which is suggested to be the formation mechanism of RSOW, are examined. The RSOW is identified and tracked as a layer with minimum potential vorticity and maximum chlorofluorocarbon-12. The pathway of the RSOW is also explored using numerical simulation. If diffusivity is not considered, the production rate of the RSOW is estimated to be 0.63 Sv using Walin’s method. By comparing this 0.63 Sv to the actual RSOW transport at the Strait of Bab el Mandeb, it is implied that the vertical diffusivity is about 3.4 x 10-5m2 s-1 . The second part of the thesis studies buoyancy-forced circulation in an idealized Red Sea. Buoyancy-loss driven circulation in marginal seas is usually dominated by cyclonic boundary currents on f-plane, as suggested by previous observations and numerical modeling. This thesis suggests that by including β-effect and buoyancy loss that increases linearly with latitude, the resultant mean Red Sea circulation consists of an anticyclonic gyre in the south and a cyclonic gyre in the north. In mid-basin, the northward surface flow crosses from the western boundary to the eastern boundary. The observational support is also reviewed. The mechanism that controls the crossover of boundary currents is further explored using an ad hoc analytical model based on PV dynamics. This ad hoc analytical model successfully predicts the crossover latitude of boundary currents. It suggests that the competition between advection of planetary vorticity and buoyancy-loss related term determines the crossover latitude. The third part of the thesis investigates three mechanisms that might account for eddy generation in the Red Sea, by conducting a series of numerical experiments. The three mechanisms are: i) baroclinic instability; ii) meridional structure of surface buoyancy losses; iii) cross-basin wind fields.

Description: This work is supported by Award Nos. USA 00002, KSA 00011 and KSA 00011/02 made by King Abdullah University of Science and Technology (KAUST) , National Science Foundation OCE0927017, and WHOI Academic Program Office.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2014

Description: Underwater acoustic communication is an extremely complex field that faces many challenges due to the time-varying nature of the ocean environment. Vector sensors are a proven technology that when utilizing their directional sensing capabilities allows us to minimize the effect of interfering noise sources. A traditional pressure sensor array has been the standard for years but suffers at degraded signal to noise ratios (SNR) and requires maneuvers or a lengthly array aperture to direction find. This thesis explores the effect of utilizing a vector sensor array to steer to the direction of signal arrival and the effect it has on equalization of the signal at degraded SNRs. It was demonstrated that utilizing a single vector sensor we were able steer to the direction of arrival and improve the ability of an equalizer to determine the transmitted signal. This improvement was most prominent when the SNR was degraded to levels of 0 and 10 dB where the performance of the vector sensor outperformed that of the pressure sensor in nearly 100% of cases. Finally, this performance improvement occured with a savings in computational expense.

Description: ONR Grants #N00014-11-10426 and #N00014-07-10738

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2014

Description: The nitrogen fixation and abundance of Trichodesmium colonies and their connections with physical processes were investigated through Video Plankton Recorder (VPR) and other observations collected in fall 2010 and spring 2011 in the western subtropical–tropical North Atlantic. A data processing procedure for estimating rare taxon abundance was devised to leverage the accuracy of manual classification and the effort savings of automatic classification. In fall 2010, local maxima in colony abundance were observed in a series of cyclones. We hypothesized Ekman transport convergence/divergence in cyclones/anticyclones as a driving mechanism and investigated the process using idealized three-dimensional models. Elevated abundances in anticyclones in spring 2011 were correlated with anomalously fresh water connected to river outflow. A bio-optical model based on carbon-normalized nitrogen fixation rates measured in fall 2010 and spring 2011 was used to estimate nitrogen fixation over the VPR transects. Mean VPR-based estimates of abundance and volume-specific nitrogen fixation rates at depth in the tropical North Atlantic were not inconsistent with estimates derived from conventional sampling methods compiled in a database by Luo et al. (2012). These findings did not reveal the systematic underestimation of deep colony populations and nitrogen fixation hypothesized by Davis and McGillicuddy (2006).

Description: This work was supported through a NASA Earth and Space Science Fellowship (NASA NNX11AL59H Understanding the role of the nitrogen- xing cyanobacterium Trichodesmium in the oceanic nitrogen and carbon cycles: in situ measurement, satellite observation, and biogeochemical modeling) as well as fellowship support from the Ocean Life Institute and Academic Programs O ce at WHOI. Additional grant support was provided by NSF OCE-0925284 Quanti cation of Trichodesmium spp. Vertical and Horizontal Abundance Patterns and Nitrogen Fixation in theWestern North Atlantic, NSF OCE-1048897 MOBY: Modeling Ocean Variability and Biogeochemical Cycles, NASA NNX13AE47G Physical and Biological Dynamics of Nonlinear Mesoscale Eddies: Satellite Observations, in situ Measurements, and Numerical Simulations on a Global Scale, and NASA NNX08AL71G Carbon cycling in the North Atlantic from regional to basin scales: satellite data, in situ observations, and numerical models.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1996

Description: This thesis addresses the question of how a highly energetic eddy field could be generated in the interior of the ocean away from the swift boundary currents. The energy radiation due to the temporal growth of non-trapped (radiating) disturbances in such a boundary current is thought to be one of the main sources for the described variability. The problem of stability of an energetic current, such as the Gulf Stream, is formulated. The study then focuses on the ability of the current to support radiating instabilities capable of significant penetration into the far-field and their development with time. The conventional model of the Gulf Stream as a zonal current is extended to allow the jet axis to make an angle to a latitude circle. The linear stability of such a nonzonal flow, uniform in the along-jet direction on a beta-plane, is first studied. The stability computations are performed for piece-wise constant and continuous velocity profiles. New stability properties of nonzonal jets are discussed. In particular, the destabilizing effect of the meridional tilt of the jet axis is demonstrated. The radiating properties of nonzonal currents are found to be very different from those of zonal currents. In particular, purely zonal flows do not support radiating instabilities, whereas flows with a meridional component are capable of radiating long and slowly growing waves. The nonlinear terms are then included in the consideration and the effects of the nonlinear interactions on the radiating properties of the solution are studied in detail. For these purposes, the efficient numerical code for solving equation for the QG potential vorticity with open boundary conditions of Orlanski's type is constructed. The results show that even fast growing linear solutions, which are trapped during the linear stage of developement, can radiate energy in the nonlinear regime if the basic current is nonzonal. The radiation starts as soon as the initial fast exponential growth significantly slows. The initial trapping of those solutions is caused by their fast temporal growth. The new mechanism for radiation is related to the nonzonality of a current.

Description: This work was supported by NSF Grant OCE 9301845.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1993

Description: There is a growing consensus that the sound generated by breaking waves is responsible for much of the ambient noise level in the ocean. While numerous field measurements have shown a strong correlation between the ambient noise spectrum level (N) in the range 100Hz to 25kHz and wind speed in the ocean, very little has been done to establish a comparable correlation between the ambient noise spectrum level and surface wave field parameters. The difficulty in establishing this relationship is remarkable given that the frequency and intensity of wave breaking are dependent on the characteristics of the wave field. In Fall 1991, an experiment was conducted from the research platform Flip 130 kilometers off the coast of Oregon, where the ambient noise between 2.5 and 25 kHz, the wind speed, and the sea surface elevation using wire wave gauges were measured. The correlation between N and the root mean square wave amplitude a was found to be poor but could be improved if the swell was filtered out from the wave elevation time series. The influence of swell on the value of a was disproportionate to the level of ambient noise since its characteristics were not directly due to the local wind-wave conditions. Observations of the dependence of the high frequency wind waves and the directional wave spectrum under turning winds suggested that the high frequency wave components responded more quickly to changes in the wind speed and wind direction than the energy-containing frequencies. The ambient noise level also correlated well with the root mean square wave slopes. This is consistent with previous laboratory measurements which showed that the steepness of a packet of waves correlates with the strength of wave breaking and with characteristics of breaking waves such as loss of momentum flux, dissipation, initial volume of air entrained, mixing, and sound generation. Comparisons of surface wave dissipation estimates using field measurements and models developed by Phillips (1985) and Hasselmann (1974) show that although the two models have very different forms, they give values that are comparable in magnitude. The relationship between the ambient noise level and log of dissipation give correlation coefficients (0.93-0.95) that are comparable to those between ambient noise and wind speed. The mean square acoustic pressure was shown to vary with the dissipation, with p2 ∝ D0.6-0.8. The results suggest that measurements of ambient sound may prove to be useful in inferring surface wave dissipation.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1997

Description: Acoustic propagation in the ocean inevitably encounters inhomogeneities of various types, which give rise to scattering. Acoustic scattering from rough water /bottom interfaces comprised of exposed rocks and sea mountains gives way to volumetric scattering in areas with fiat interfaces and thick sediment cover. The data analysis of the ARSRP backscattering experiment revealed that random inhomogeneities in two irregular layers beneath the seafloor were the primary contributors to oblique backscattering in a sediment pond on the western flank of the Mid-Atlantic Ridge. In this thesis, an attempt has been made to model monostatic backscattering from 3-D volume inhomogeneities in the sediment and to compare the results with the ARSRP backscattering data. A scattering process cannot be modeled correctly without a proper account of the incident field . Several approximate propagation models have been evaluated against the exact solution, while the appropriateness of using the equivalent surface scattering strength in volume scattering characterizations is studied. This study concludes that precautions need to be taken in modeling both the propagation effects and the scattering mechanisms associated with the bottom volume scattering process. A volume scattering model based on perturbation theory and the Born approximation is developed incorporating contributions from both sound speed and density fluctuations. With the propagation part handled accurately by OASES and random fluctuations generated effectively by a new scheme modified from the spectral method, the model is capable of simulating the monostatic backscattered field and time series due to 3-D volumetric sediment inhomogeneities. Both the characteristic length scale and power spectrum descriptions of the random inhomogeneities are shown to have great impact on the backscattered field by parameter studies in a free-space scenario. The important roles played by horizontal anisotropy and the vertical correlation of the random field have been demonstrated. Density fluctuations are further confirmed to be the dominant force in backscattering. The model matches the ARSRP backscattering data very well, with the fluctuations of sound speed and density in the two irregular layers described by a power law type of power spectrum.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1997

Description: This work is concerned with coherent communication by means of acoustic signals over underwater communication channels. The estimated scattering functions of real data ranging from the Arctic environment to tropical waters show that underwater communication channels can not be captured by a single, simple channel model. This thesis considers mainly a subset of underwater communication channels where the Doppler spread is more severe than the delay spread. An appropriate representation of the linear time- variant channel is introduced, and the wide sense stationary uncorrelated scattering (WSSUS) channel assumption enables characterization in terms of scattering functions. The concept of Doppler lines, which are frequency domain filters, is used in the derivation of a receiver for Doppler spread channels. The channel is simulated by means of a ray representation for the acoustic field and a time-variant FIR filter. The impact of physical ocean processes on the Doppler spread is demonstrated, and from this modeling explanations for the Doppler spread observed in real data are obtained. A decision feedback equalizer (DFE) adapted with recursive least squares (RLS) is analyzed, and its limit with respect to pure Doppler spread is found. By using the DFE with a phase locked loop (PLL) suboptimal system behavior is found, and this is verified on real data. In the case of a simple Doppler shift the cross-ambiguity function is used to estimate the shift, and the received signal is phase rotated to compensate this before it enters the receiver. A modified RLS called the time updated RLS (TU-RLS) is presented, and it is used in a new receiver. This receiver is initialized by means of the cross-ambiguity function and the performance is characterized by probability of decoding error vs delay spread, Doppler spread and SNR. The receiver uses Doppler lines to compensate both discrete and continuous Doppler spread. The receiver stability depends on the conditioning of the block diagonal correlation matrix propagated by the TU-RLS. The receiver is used to decode both real and simulated data, and some of these data are severely Doppler spread.

Description: The work in this thesis was funded in part by the Norwegian Research Counsil under a general PhD contract, the education office of Woods Hole Oceanographic Institution and the Office of Naval Research under contracts N00014-95-0153 and N00014-95-l-0322.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1997

Description: As part of the Shallow Water Acoustics in a Random Medium (SWARM) experiment [1], a sixteen element WHOI vertical line array (WVLA) was moored in 70 meters of water off the New Jersey coast. This array was sampled at 1395 Hz or higher for the seven days it was deployed. Tomography sources with carrier frequencies of 224 and 400 Hz were moored about 32 km shoreward, such that the acoustic path was anti-parallel to the primary propagation direction for shelf generated internal wave solitons. Two models for the propagation of normal modes through a 2-D waveguide with solitary internal wave (soliton) scattering included are developed to help in understanding the very complicated mode arrivals seen at the WVLA. The simplest model uses the Preisig and Duda [2] sharp interface approximation for solitons, allowing for rapid analysis of the effects of various numbers of solitons on mode arrival statistics. The second model, using SWARM thermistor string data to simulate the actual SWARM waveguides, is more realistic, but much slower. The analysis of the actual WVLA data yields spread, bias, wander, and intensity fluctuation signals that are modulated at tidal frequencies. The signals are consistent with predicted relationships to the internal wave distributions in the waveguides.

Description: The funds for my education were provided by the Office of Naval Research through an ONR Fellowship (MIT award 002734-001); the funds for SWARM were also provided by the Office of Naval Research through ONR Grant N00014-95-0051.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2013

Description: Two high-resolution mooring arrays extending from the outer shelf to the mid continental slope are used to elucidate shelf-basin exchange at the inflow to and the outflow from the Arctic Ocean. Pacific Water entering the Arctic Ocean forms the Western Arctic shelfbreak current along the Beaufort Sea slope. Data from the mooring array at 152°W—approximately 150 km east of Pt. Barrow, AK—reveals that this current has two distinct states in summer depending on the water mass it transports. When advecting Alaskan Coastal Water it is surface-intensified and both baroclinically and barotropically unstable. This configuration lasts about a month with an average transport of 0.7 Sv. When advecting Chukchi Summer Water the current is bottom-intensified and is only baroclinically unstable. This state also exists for approximately a month with an average transport of 0.6 Sv. The strong mean-to-eddy energy conversion causes both configurations of the current to spin down over a distance of a few hundred kilometers, suggesting that warm Pacific Water does not enter the Canadian Arctic Archipelago via this route. Dense water formed in the Nordic Seas overflows the Denmark Strait and undergoes vortex stretching, forming intense cyclones that propagate along the East Greenland slope. Data from the mooring array at 65°N—roughly 300 km downstream of Denmark Strait—was used to determine the full water column structure of the cyclones. On average a cyclone passes the array every other day in the vicinity of the 900 m isobath, although the depth range of individual cyclones ranges between the 500 m and 1600 m isobaths. The cyclones self-propagate at 0.45 m/s and are also advected by the mean flow of 0.27 m/s, resulting in a total propagation speed of 0.72 m/s. They have a peak azimuthal speed of 0.22 m/s at a radius of 7.8 km and contain overflow water in their core. In the absence of the cyclones, the background flow is dominated by the East Greenland Spill Jet. This is shown to be a year-round feature transporting 2–4 Sv of dense water equatorward along the upper continental slope.

Description: Financial support for this work was provided by National Science Foundation grants OCE-0726640 and OCE-0612143, by the Arctic Research Initiative at WHOI, by the Y-S Anonymous Fellowship from the Office of the Dean of Graduate Education at MIT, and by WHOI Academic Programs Office funds.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution December 1996

Description: The transformation of potential vorticity within and stability of nonlinear deep western boundary currents in an idealized tropical ocean are studied using a shallowwater model. Observational evidence indicates that the potential vorticity of fluid parcels in deep western boundary currents must change sign as they cross the equator, but this evidence is otherwise unable to clarify the process. A series of numerical experiments investigate this transformation in a rectangular basin straddling the equator. A mass source located in the northwestern corner feeds fluid into the domain where it is constrained to cross the equator to reach a distributed mass sink. Dissipation is included as momentum diffusion. The Reynolds number, defined as the ratio of the mass source per unit depth to the viscosity, determines the nature of the flow, and a critical value, Rec, divides its possible behavior into two regimes. For Re 〈 Rec, the flow is laminar and well described by linear theory. For Re just above the critical value, the flow is time-dependent, with cyclonic eddies forming in the western boundary current near the equator. For still larger Reynolds number, eddies of both signs emerge and form a complicated, interacting network that extends into the basin several deformation radii from the western boundary, as well as north and south of the equator. The eddy field is established as the mechanism for potential vorticity transformation in nonlinear cross-equatorial flow. The analysis of vorticity fluxes follows from the flux-conservative form of the absolute vorticity equation. It is shown that the zonally integrated meridional flux of vorticity across the equator using no slip boundary conditions is virtually zero even in the strongly nonlinear limit suggesting that the eddies are extremely efficient vorticity transfer agents. A decomposition of the vorticity fluxes into components due to mean advection, eddy transport, and friction, reveals the growth with Reynolds number of a turbulent boundary layer that exchanges vorticity between the inertial portion of the boundary current and a frictional sub-layer where modification is straightforward. A linear stability analysis of the shallow-water system in the tropical ocean examines the initial formation of the eddy field. The formulation assumes that the basic state is purely meridional and on a local f-plane. Realistic western boundary current profiles undergo a horizontal shear instability that is partially stabilized by viscosity. Calculations at several latitudes indicate that the instability is enhanced in the tropics where the internal deformation radius is a maximum. The linear stability analysis predicts a length scale of the disturbance, a location for its origin, and a critical Reynolds number that agree well with numerical results.

Description: Financial support for this research was provided by NSF grant number OCE- 9115915 and ONR ASSERT grant number N00014-94-1-0844.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1996

Description: This thesis addresses the issue, "Which approach to instabilities-temporal, spatial or pulse theory- is the most appropriate model for the Gulf Stream?" I also address the question of how the observations might be compared to theory. This thesis consists of two closely related parts: analytical studies that compare the three types of instability using the same realistic velocity and topography profiles; and numerical modeling that uses a continuous forcing function to examine the three types of theory in the direct context of the Gulf Stream.

Description: My first three years in the Joint Program were supported by the National Science Foundation under grant OCE-9011066 and last two and half years under NSF grant OCE-9314140.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2012

Description: Adaptive equalization is an important aspect of communication systems in various environments. It is particularly important in underwater acoustic communication systems, as the channel has a long delay spread and is subject to the effects of time- varying multipath fading and Doppler spreading. The design of the adaptation algorithm has a profound influence on the performance of the system. In this thesis, we explore this aspect of the system. The emphasis of the work presented is on applying concepts from inference and decision theory and information theory to provide an approach to deriving and analyzing adaptation algorithms. Limited work has been done so far on rigorously devising adaptation algorithms to suit a particular situation, and the aim of this thesis is to concretize such efforts and possibly to provide a mathematical basis for expanding it to other applications. We derive an algorithm for the adaptation of the coefficients of an equalizer when the receiver has limited or no information about the transmitted symbols, which we term the Soft-Decision Directed Recursive Least Squares algorithm. We will demonstrate connections between the Expectation-Maximization (EM) algorithm and the Recursive Least Squares algorithm, and show how to derive a computationally efficient, purely recursive algorithm from the optimal EM algorithm. Then, we use our understanding of Markov processes to analyze the performance of the RLS algorithm in hard-decision directed mode, as well as of the Soft-Decision Directed RLS algorithm. We demonstrate scenarios in which the adaptation procedures fail catastrophically, and discuss why this happens. The lessons from the analysis guide us on the choice of models for the adaptation procedure. We then demonstrate how to use the algorithm derived in a practical system for underwater communication using turbo equalization. As the algorithm naturally incorporates soft information into the adaptation process, it becomes easy to fit it into a turbo equalization framework. We thus provide an instance of how to use the information of a turbo equalizer in an adaptation procedure, which has not been very well explored in the past. Experimental data is used to prove the value of the algorithm in a practical context.

Description: Support from the agencies that funded this research- the Academic Programs Office at WHOI and the Office of Naval Research (through ONR Grant #N00014-07-10738 and #N00014-10-10259).

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1998

Description: Work on the forward problem in zooplankton bioacoustics has resulted in the identification of three categories of acoustic scatterers: elastic-shelled (e.g. pteropods), fluid-like (e.g. euphausiids), and gas-bearing (e.g. siphonophores). The relationship between backscattered energy and animal biomass has been shown to vary by a factor of —19,000 across these categories, so that to make accurate estimates of zooplankton biomass from acoustic backscatter measurements of the ocean, the acoustic characteristics of the species of interest must be well-understood. This thesis describes the development of both feature based and model based classification techniques to invert broadband acoustic echoes from individual zooplankton for scatterer type, as well as for particular parameters such as animal orientation. The feature based Empirical Orthogonal Function Classifier (EOFC) discriminates scatterer types by identifying characteristic modes of variability in the echo spectra, exploiting only the inherent characteristic structure of the acoustic signatures. The model based Model Parameterisation Classifier (MPC) classifies based on correlation of observed echo spectra with simplified parameterisations of theoretical scattering models for the three classes. The Covariance Mean Variance Classifiers (CMVC) are a set of advanced model based techniques which exploit the full complexity of the theoretical models by searching the entire physical model parameter space without employing simplifying parameterisations. Three different CMVC algorithms were developed: the Integrated Score Classifier (ISC), the Pairwise Score Classifier (PSC) and the Bayesian Probability Classifier (BPC); these classifiers assign observations to a class based on similarities in covariance, mean, and variance, while accounting for model ambiguity and validity. These feature based and model based inversion techniques were successfully applied to several thousand echoes acquired from broadband (-350 kHz - 750 kHz) insonifications of live zooplankton collected on Georges Bank and the Gulf of Maine to determine scatterer class. CMVC techniques were also applied to echoes from fluid-like zooplankton (Antarctic krill) to invert for angle of orientation using generic and animal-specific theoretical and empirical models. Application of these inversion techniques in situ will allow correct apportionment of backscattered energy to animal biomass, significantly improving estimates of zooplankton biomass based on acoustic surveys.

Description: Thanks to the WHOI/MIT Joint Program Education Office for partial funding. Other sources of funding for my thesis work include the Ocean Acoustics, Oceanic Biology and URIP programs of the Office of Naval Research grant numbers N00014-89- J-1729, N00014-95-1-0287 and N00014-924-1527, and the Biological Oceanography program of the National Science Foundation grant number OCE-9201264.

Description: Submitted in partial fulfillment of the requirements for the degree of Ocean Engineer at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution August 1989

Description: When ocean waves in deep water interact with a current, the direction of propagation and characteristics of the waves such as height and length are affected. Swell in the open ocean can undergo significant refraction as it passes through major current systems like the Gulf Stream or Antarctic Circumpolar Current. Remote sensing techniques such as synthetic aperture radars (SAR) have the potential to detect wave systems over a wide geographical area. Combining a model for wave refraction in the presence of currents with SAR measurements, the inverse problem of using the measured wave data can be solved to determine the direction and magnitude of the intervening currents. In this study the behavior of swell measured by SAR on a satellite pass over the Gulf Stream is examined. The refraction predicted by a numerical model under conditions of varying current profiles and velocities is compared to SAR generated wave spectra. By matching the current profile which results in the best correlation of wave refraction to the SAR data, the tomographic problem of measuring the Gulf Stream current is solved. The best correlation between the model and SAR data is obtained when a current is modeled by a top hat velocity profile with a direction of 75° and a current speed of 2 m/s. The direction agrees with that visually observed from the SAR images, and the direction and speeds are close to the Coast Guard estimates for the Gulf Stream at the time of the SEASAT,pass. The current profiles used did not take into account a possible widening of the Gulf Stream at the position of the satellite overpass. There is a great deal of scatter in the SAR data, both before and in the Gulf Stream, so it is difficult to correlate every point with specific current behavior, but the increase in wave length and change in wave angle in the center of the Gulf Stream seem to indicate that there may be a non-uniform feature such as the formation of an eddy or other lateral variability near the current's edge.

Description: I was supported by the U. S. Navy.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1990

Description: Travel-times of acoustic signals were measured between a bottom-mounted source near Oahu and four bottom-mounted receivers located near Washington, Oregon, and California in 1988 and 1989. This paper discusses the observed tidal signals. At three out of four receivers, observed travel times at M2 and S2 periods agree with predictions from barotropic tide models to within ±30° in phase and a factor of 1.6 in amplitude. The discrepancy at the fourth receiver can be removed by including predicted effects of phase-locked baroclinic tides generated by seamounts. Our estimates of barotropic M2 tidal dissipation by seamounts vary between 2 x 1016 and 1 X 1018 erg·s-1. The variation by two orders of magnitude is due to uncertainties in the numbers and sizes of seamounts. The larger dissipation (1 x 1018 erg·s-1) is the same order as previous estimates and amounts to 4% of the total dissipation at M2.

Description: Submitted in partial fulfillment of the requirements for the degree of Ocean Engineer at the Massachusetts Institute of Technology and Woods Hole Oceanographic Institution January 1994

Description: In 1991 the Heard Island Feasibility Test demonstrated that it is possible to transmit coded acoustic signals nearly half way around the world. One of the key issues in the feasibility test was to determine the spatial structure of the received transmissions. In this thesis, data from the Canadian Defense Research Establishment Pacific horizontal line array is used to form an estimate of the directional power spectrum. This spectrum determines if any horizontal multipath is detectable. The preliminary signal conditioning, including frequency spectrum estimation and demodulation required before beamforming is described. Conventional and adaptive beamforming methods are examined with synthetic data to demonstrate the limitations on the directional spectrum results. The principle result of this work is that no stable horizontal multipath is evident. The mean arrival angle for the five hours of data analyzed is 212° ± 1.5°.

Description: The Office of Naval Research provided funding for the author under the ONR fellowship program.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2012

Description: Interactions between the ocean circulation in sub-ice shelf cavities and the overlying ice shelf have received considerable attention in the context of observed changes in flow speeds of marine ice sheets around Antarctica. Modeling these interactions requires parameterizing the turbulent boundary layer processes to infer melt rates from the oceanic state at the ice-ocean interface. Here we explore two such parameterizations in the context of the MIT ocean general circulation model coupled to the z-coordinates ice shelf cavity model of Losch (2008). We investigate both idealized ice shelf cavity geometries as well as a realistic cavity under Pine Island Ice Shelf (PIIS), West Antarctica. Our starting point is a three-equation melt rate parameterization implemented by Losch (2008), which is based on the work of Hellmer and Olbers (1989). In this form, the transfer coefficients for calculating heat and freshwater fluxes are independent of frictional turbulence induced by the proximity of the moving ocean to the fixed ice interface. More recently, Holland and Jenkins (1999) have proposed a parameterization in which the transfer coefficients do depend on the ocean-induced turbulence and are directly coupled to the speed of currents in the ocean mixed layer underneath the ice shelf through a quadratic drag formulation and a bulk drag coefficient. The melt rate parameterization in the MITgcm is augmented to account for this velocity dependence. First, the effect of the augmented formulation is investigated in terms of its impact on melt rates as well as on its feedback on the wider sub-ice shelf circulation. We find that, over a wide range of drag coefficients, velocity-dependent melt rates are more strongly constrained by the distribution of mixed layer currents than by the temperature gradient between the shelf base and underlying ocean, as opposed to velocity-independent melt rates. This leads to large differences in melt rate patterns under PIIS when including versus not including the velocity dependence. In a second time, the modulating effects of tidal currents on melting at the base of PIIS are examined. We find that the temporal variability of velocity-dependent melt rates under tidal forcing is greater than that of velocity-independent melt rates. Our experiments suggest that because tidal currents under PIIS are weak and buoyancy fluxes are strong, tidal mixing is negligible and tidal rectification is restricted to very steep bathymetric features, such as the ice shelf front. Nonetheless, strong tidally-rectified currents at the ice shelf front significantly increase ablation rates there when the formulation of the transfer coefficients includes the velocity dependence. The enhanced melting then feedbacks positively on the rectified currents, which are susceptible to insulate the cavity interior from changes in open ocean conditions.

Description: National Science and Engineering Research Council of Canada

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution May 1989

Description: The problem of a low-frequency acoustic plane wave incident upon a free surface coupled to a semi-infinite elastic plate surface, is solved using an analytic approach based on the Wiener-Hopf method. By low-frequency it is meant that the elastic properties of the plate are adequately described by the thin plate equation (kH ≲ 1). The diffraction problem relates to issues in long range sound propagation through partially ice-covered Arctic waters, where open leads or polynya on the surface represent features from which acoustic energy can be diffracted or scattered. This work focusses on ice as the material for the elastic plate surface, and, though the solution methods presented here have applicability to general edge diffraction problems, the results and conclusions are directed toward the ice lead diffraction process. The work begins with the derivation of an exact solution to a canonical problem: a plane wave incident upon a free surface (Dirichlet boundary condition) coupled to a perfectly rigid surface (Neumann boundary condition). Important features of the general edge diffraction problem are included here, with the solution serving as a guideline to the more complicated solutions presented later involving material properties of the boundary. The ice material properties are first addressed using the locally reacting approximation for the input impedance of an ice plate, wherein the effects of elasticity are ignored. This is followed by use of the thin plate equation to describe the input impedance, which incorporates elements of elastic wave propagation. An important issue in working with the thin plate equation is the fluid loading pertaining to sea ice and low-frequency acoustics, which cannot be characterized by simplifying heavy or light fluid loading limits. An approximation to the exact kernel of the Wiener-Hopf functional equation is used here, which is valid in this mid-range fluid loading regime. Use of this approximate kernel allows one to proceed to a complete and readily interpretable solution for the far field diffracted pressure, which includes a subsonic flexural wave in the ice plate. By using Green's theorem, in conjunction with the behavior of the diffracted field along the two-part planar boundary, the functional dependence of ∏D (total diffracted power) in terms of k (wavenumber), H (ice thickness), α (grazing angle) and the combined elastic properties of the ice sheet and ambient medium, is determined. A means to convert ∏D into an estimate of dB loss per bounce is developed using ray theoretical methods, in order to demonstrate a mechanism for acoustic propagation loss attributed directly to ice lead diffraction effects. Data from the 1984 MIZEX (Marginal Ice Zone Experiments) narrow-band acoustic transmission experiments are presented and discussed in this context.

Description: I also gratefully acknowledge financial support provided by the WHOI Education Office and the Office of Naval Research.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1989

Description: Variability of the Florida Current has been monitored via acoustic tomography. A reciprocal tomography experiment was conducted in the eastern half of the Florida Straits during mid October and November, 1983. A triangular array of transceivers, with leg separations of approximately 45 kilometers, was deployed at 27°N. The presence of a surface mixed layer in the region allowed for the ducted propagation of acoustic energy in the surface layer. A deeper layer was sampled by an unresolved group of refracted, bottom reflected ray arrivals. Incorporating the complete set of arrivals, we are able to obtain depth dependent estimates of the temperature field, current velocity, and relative vorticity. The oceanography of the region has been shown to be dominated by the lateral shifting of the surface and subsurface core of the Florida Current. The influx of westward flowing water through the Northwest Providence Channel at 26°N also appears as a large scale signal in the eastern Florida Straits. Low frequency fluctuations of temperature, current velocity, and vorticity occur at periods ranging from several days to nearly two weeks, and are intimately related to meandering of the Florida Current system.

Description: This research was carried out under ONR contract N00014-86-K-0751.

Description: Submitted in partial fulfillment of the requirements for the degree of Ocean Engineer at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1990

Description: The propagation of low frequency seismo-acoustic waves in the Arctic Ocean ice canopy is examined through the analysis of hydrophone and geophone data sets collected in 1987 at an ice camp designated PRUDEX in the Beaufort Sea. Study of the geophone time series generated by under-ice explosive detonations reveals not only the expected longitudinal and flexural waves in the ice plate, but also an unexpected horizontally-polarized transverse (SH) wave arriving at a higher amplitude than the other wave types. The travel paths of all three observed wave types are found to be refracted in the horizontal plane along a line coincident with a known ridge separating the ice canopy locally into two distinct half-plates, the first of thin first year ice and the second of thicker multi-year ice. The origin of the SH wave appears to be near the detonation and not associated with the interaction of longitudinal, flexural or waterborne waves with the ridge line. The need to determine the exact location of each detonation from the received time series highlights the dramatic superiority of geophones over hydrophones in this application, as does the ability to detect the anomalous SH waves and the refracted ray paths, neither of which are visible in the hydrophone data. Inversion of the geophone data sets for the low frequency elastic parameters of the ice is conducted initially by treating the ice as a single homogeneous isotropic plate to demonstrate the power of SAFARI numerical modeling in this application. A modified stationary phase approach is then used to extend SAFARI modeling to invert the data sets for the elastic parameters of the two ice half-plates simultaneously. The compressional/shear bulk wave speeds estimated in the half-plates, 3500/1750 m/s in the multi-year ice and 3000/1590 m/ s in the new ice, are comparable to previously obtained values; however, the compressional/shear attenuation values in the two half-plates, 1.0/2.99 d/Bλ. and 1.0/2.67 dB/λ, respectively, are somewhat greater than previously measured values and four times greater than estimates extrapolated from high frequency data.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1989

Description: A triangular CTD/ADCP survey was made across the Kuroshio west of Kyushu aboard the R/V Thompson during January, 1986 in order to investigate the water properties and flow field in the Kuroshio. A similar CTD survey was made in July, 1986 aboard the R/V Washington to study the seasonal variability in the Kuroshio. The Kuroshio in this region exhibited a marked seasonal change in its near-surface stratification and water properties. In January, the Kuroshio water was separated from the vertically well-mixed coastal water over the shelf by a strong front located near the shelf break. Horizontal mixing between the Kuroshio and coastal water was observed but was limited near the shelf break. In July, surface coastal water extended far past the shelf break over the Kuroshio region near the surface, and in turn, Kuroshio water intruded onto the shelf near the bottom. Mixing between the Kuroshio and coastal water was found over much of the mid and outer shelf and upper slope, spanning a cross-stream distance of 75 km. In addition, evidence of deep vertical mixing within the Kuroshio itself was found near 32.0°N and 128.2°E, most likely due to internal tidal mixing over the slope. Since Loran C navigation coverage in the study region was poor during the R/V Thompson cruise, a simple averaging technique has been used to convert the ADCP data into an absolute velocity. An error analysis shows that the total error in the absolute ADCP velocity was about ±5 cm/s. The absolute geostrophic velocity using the absolute Doppler velocity at 60 m as the reference velocity was then calculated for the sides of the triangle. The results show that the ADCP velocity shear was in good agreement with the geostrophic shear in the Kuroshio. The Kuroshio flowed through the western section as a coherent current, then split into two streams around a tall seamount as it left through the eastern section. Some recirculation also occurred between the core of the Kuroshio and the slope as well as near the seamount. The geostrophic velocity field calculated relative to the bottom missed some of the important features of the true flow field such as splitting of the Kuroshio and the recirculation in the slope region. The volume, salt and heat transports of the Kuroshio during the January 1986 survey have been cakulated using the absolute geostrophic velocity and CTD data. The volume transport of the Kuroshio west of Kyushu in January 1986 was 31.7± 2.0 Sv, which is comparable to that of the Gulf Stream in the Florida Strait. The volume transport through the triangle was conserved within measurement uncertainty, so that a streamfuction field can be defined by the transport. The resulting streamlines clearly show the structure of the flow field in the Kuroshio and its adjacent currents during the survey. The advective heat transport of the Kuroshio west of Kyushu in January 1986 was 28.2 ± 1.8 x 1014 W. The salt transport in January 1986 was about 108.0 ± 7.3 x 1010 kg/s, and the net salt flux was zero within measurement error. Analysis of the potential vorticity based on the January 1986 absolute geostrophic velocity field shows that the total potential vorticity in the Kuroshio may be approximately given by the product of the vertical gradient of the potential density and the sum of the planetary and relative vorticities. The distribution of relative vorticity plays a significant role in determining the structure of the potential vorticity in the Kuroshio. The path of the Kuroshio can be traced in the field of potential vorticity. Facing in the direction of the current, the axis of the maximum velocity is located to the right of the core of maximum potential vorticity. Finally, the Kuroshio was potentially unstable since the gradient of potential vorticity changed its sign on potential density surfaces across the Kuroshio.

Description: Submitted in partial fulfillment of the requirements for the degree of Ocean Engineer at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1990

Description: Pulse-like acoustic signals are transmitted from an acoustic source near Oahu to seven receivers off the west coast of the United States for a 124-day period in 1988. Acoustic travel-time oscillations are observed in the received signal at periods between 15 and 23 hours, which are caused by barotropic (or first or second mode baroclinic) flu ctuations in the ocean. It is shown that these fluctuations cannot be local processes isolated to either the source or to the receivers. It is further shown that resonant barotropic gravity wave modes (Platzman et al., 1981) are not consistent with the data. The cause of these flu ctuations remains unresolved, but the data and other oceanographic measurements put many constraints on the process causing these fluctuations.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1996

Description: The shallow water acoustic channel supports far-field propagation in a discrete set of modes. Ocean experiments have confirmed the modal nature of acoustic propagation, but no experiment has successfully excited only one of the suite of mid-frequency propagating modes propagating in a coastal environment. The ability to excite a single mode would be a powerful tool for investigating shallow water ocean processes. A feedback control algorithm incorporating elements of adaptive estimation, underwater acoustics, array processing and control theory to generate a high-fidelity single mode is presented. This approach also yields a cohesive framework for evaluating the feasibility of generating a single mode with given array geometries, noise characteristics and source power limitations. Simulations and laboratory waveguide experiments indicate the proposed algorithm holds promise for ocean experiments.

Description: Josko Catipovic funded my research for summer of 1992 on the Office of Naval Research Grant Number N00014-92-J-1661 and from June 1993 through August 1995 on Defense Advanced Research Projects Agency Grant Number MDA972-92-J- 1041. The Office of Naval Research Grant N00014-95-1-0362 to MIT supported the computer facilities used to do much of this work.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1992

Description: This thesis consists of two parts: (I) variability of currents and water properties in late spring in the northern Great South Channel and (II) numerical study of stratified tidal rectification over Georges Bank. In part I, the combined analysis of CTD, ADCP, satellite-track drifters, and zooplankton distributions in the northern Great South Channel clearly shows (1) the seasonal evolution of the surface low salinity plume, (2) the threedimensional structure of residual flow, and (3) a coherent relationship between the surface low salinity plume and high concentration of zooplankton. In part II, the numerical model of Georges Bank shows that as the fluid becomes stratified , tidal mixing and rectification intensify both along- and cross-bank residual currents and modify the vertical structure of the flow. Along- and cross-bank residual currents increase as either stratification increases or the depth of the bank decreases. Model results over Georges Bank are in good agreement with observation, particularly in the position of the tidal mixing front and residual currents on the northern flank of the Bank.

Description: This research was supported by the National Science Foundation under grants OCE 87-13988 and OCE 91-01034 and by the National Center for Atmospheric Research (NCAR) under computer time grants Nos. 35781029 and 35781035.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1998

Description: The goal of this thesis is to develop a methodology to interpret sound scattered from the seafloor in terms of seafloor structure and subseafloor geological properties. Specifically, this work has been directed towards the interpretation of matched-filtered, beamformed monostatic acoustic reverberation data acquired on the west flank of the Mid-Atlantic Ridge when the seafloor is insonified by a band-limited, low-grazing-angle acoustic pulse. This research is based on the hypothesis that observed backscatter signals are produced by a combination of seafloor (interface) scattering and subseafloor (volume) scattering from structure having variations at scale lengths similar to the wavelength of the insonifying acoustic field. Analysis of monostatic reverberation data acquired during the Site A experiment (Run 1) of the Acoustic Reverberation Special Research Program 1993 Acoustics Cruise suggests that the scattered signals cannot be accounted for quantitatively in terms of large-scale slope, even though a strong correspondence between high intensity backscatter and seafloor ridges is observed. In order to investigate and quantify the actual sources of seafloor scattering, a numerical modeling study of seafloor models is undertaken using a finite-difference solution to the elastic wave equation. Geological data available at Site A and published reports describing geological properties of similar deep ocean crustal regions are used to develop a realistic seafloor model for the study area with realistic constraints on elastic parameters. Wavelength-scale heterogeneity in each model, in the form of seafloor roughness and subseafloor volume heterogeneity is defined using stochastic distributions with Gaussian autocorrelations. These distributions are quantified by their correlation lengths and standard deviation in amplitude. In order to incorporate all seafloor structure in a single parameterization of seafloor scattering, large-scale slope and wavelength-scale seafloor spatial parameters (rms height and correlation length), are included, along with the acoustic beam grazing-angle relative to a horizontal seafloor, in the definition of an 'effective grazing angle'. The Rayleigh roughness parameter, which depends on grazing angle of the insonification, is then redefined using the effective grazing angle and calculated for a variety of seafloor models. Scattering strengths are shown to vary systematically but nonlinearly with the 'effective Rayleigh roughness parameters' of horizontal rough seafloor models. This leads to an approximate interpretation scheme for backscatter intensity. In general, variation in backscattering is found to be dominated by the scattering from rough seafloor. If the seafloor is smooth or very low velocity (e.g., sediment), then scattering from volume heterogeneity becomes an important factor in the backscattered field. Both wavelength-scale seafloor roughness and volume heterogeneity are shown to be capable of producing the levels of variation in intensity observed in monostatic reverberation experiments. Variations in large-scale seafloor slope and subseafloor average velocity are shown to influence the backscatter response of seafloor models.

Description: My first year of study at MIT was supported by a Chevron Fellowship. The work in this thesis was funded by the Office of Naval Research under grant numbers N00014-93-1-1352, N00014-90-J-1493, and N00014-95-1-0506.

Description: Submitted in partial fulfillment of the requirements for the degree of Ocean Engineer at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution May 1995

Description: The data analysis of a deep-sea bottom backscattering experiment, carried out over a sediment pond on the western flank of the Mid-Atlantic Ridge in July 1993 with a 250- 650 Hz chirp source and a vertical receiving array suspended near the fiat seafloor, is presented in this thesis. Reflected signals in the normal incidence direction as the output of endfire beamforming are used to determine the sediment structure. The sediment is found to be horizontally stratified, except for two irregular regions, each about 20 m t hick, located around 18 m and 60 m beneath the water-sediment interface. Multiple constraints beamforming is shown to be effective in removing coherent reflections from internal stratified layers, which is critical to the analysis of bottom backscattering. With backscattered signals obtained by beamforming, the above-mentioned two inhomogeneous regions are found to be the dominant factors on the bottom backscattered field, both in the normal incidence and oblique directions. The backscattering strength as a function of grazing angle is estimated for each of the two regions.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution March 1990

Description: This thesis introduces a new procedure for the enhancement of acoustic images of the bottom of the sea produced by side-scan sonars. Specifically, it addresses the problem of estimating and correcting geometric distortions frequently observed in such images as a consequence of motion instabilities of the sonar array. This procedure estimates the geometric distortions from the image itself, without requiring any navigational or attitude measurements. A mathematical model for the distortions is derived from the geometry of the problem, and is applied to estimates of the local degree of geometric distortion obtained by cross-correlating segments of adjacent lines of the image. The model parameters are then recursively estimated through deterministic least-squares estimation. An alternative approach based on adaptive Kalman filtering is also proposed, providing a natural framework in which a priori information about the array dynamics may be easily incorporated. The estimates of the parameters of the distortion model are used to rectify the image, and may also be used for estimating the attitude parameters of the array. A simulation is employed to evaluate the effectiveness of this technique and examples of its application to high-resolution side-scan sonar images are provided.

Description: This work was produced under sponsorship of the Conselho Nacional de Desenvolvimento Cient{jico e Tecnol6gico (CNPq), an agency of the Government of the Federative Republic of Brazil, and was supported in part by the Defense Advanced Research Projects Agency monitored by the Office of Naval Research under Grant No. N00014-89-J-1489, in part by the National Science Foundation under Grant No. MIP 87-14969, and in part by Sanders Associates, Incorporated.

Description: Submitted in partial fulfillment of the requirements for the degree of Ocean Engineer at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1991

Description: A model is developed for the prediction of the seismo-acoustic noise spectrum in the microseism peak region (0.1 to 0.7 Hz). The model uses a theory developed by Cato [J. Acoust. Soc. Am., 89 , 1096-1112 (1991)] for an infinite depth ocean in which the surface orbital motion caused by gravity waves may produce acoustic waves at twice the gravity wave frequency. Using directional wave spectra as inputs, acoustic source levels are computed and incorporated into a more realistic environment consisting of a horizontally stratified ocean with an elastic bottom. Noise predictions are made using directional wave spectra obtained from the SWADE surface buoys moored off the coast of Virginia and the SAFARI sound propagation code, with a bottom model derived using wave speeds measured in the EDGE deep seismic reflection survey. The predictions are analyzed for noise level variations with frequency, wave height, wind direction, and receiver depth. These predictions are compared to noise measurements made in ECONOMEX using near-bottom receivers located close to the surface buoys. Good agreement is found between the predictions and observations under a variety of environmental conditions.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1992

Description: An ultra-short baseline acoustic navigation system has been developed which is capable of determining bearing to a sound source with an error of less than 1° in typical operational conditions. The system has a demonstrated ability to operate in an environment in which multipath interference is significant. A DSP microprocessor is used to process signals received by two hydrophones from a 26 kHz toneburst sound source. This processing power is used to implement features not commonly available with commercial systems. The system has the capability to make an on-line measurement of the signal-tonoise ratio, which can be used to estimate the confidence which should be placed in the data. Estimates of phase difference and signal power are generated many times within each received pulse, so the effects of multi path interference throughout the pulse can be observed. Results of tests at several ranges are presented, and compared to performance models developed in the thesis. System performance is quantified, and an effort is made to understand the effects of multipath arrivals.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1993

Description: The dynamical aspects involved in the assimilation of altimeter data in a numerical ocean model have been investigated. The model used for this study is a quasi-geostrophic model of the Gulf Stream region. The data that have been assimilated are maps of sea surface height which have been obtained as the superposition of sea surface height variability deduced from the Geosat altimeter measurements and a mean field constructed from historical hydrographic data. The method used for assimilating the data is the nudging technique. Nudging has been implemented in such a way as to achieve a high degree of convergence of the surface model fields toward the observations. We have analyzed the mechanisms of the model adjustment, and the final statistical equilibrium characteristics of the model simulation when the surface data are assimilated. Since the surface data are the superposition of a mean component and an eddy component, in order to understand the relative role of these two components in determining the characteristics of the final st atistical steady state, we have considered two different experiments: in the first experiment only the climatological mean field is assimilated, while in the second experiment the total surface streamfunction field (mean + eddies) has been used. We have found that the mean component of the surface data determines, to a large extent, the structure of the flow field in the subsurface layers, while the eddy field, as well as the inflow/outflow conditions at the open boundaries, affect its intensity. In particular, if surface eddies are not assimilated only a weak flow develops in the two deeper model layers where no inflow/ outflow is prescribed at the boundaries. Comparisons of the assimilation results with available in situ observations show a considerable improvement in the degree of realism of the climatological model behavior, with respect to the model in which no data are assimilated. In particular, the possibility of building into the model more realistic eddy characteristics, through the assimilation of the surface eddy field, proves very successful in driving components of the mean model circulation that are in good agreement with the available observations.

Description: This research was carried out with the support of the National Aeronaut ics Space Administration, through a contract to MIT from the Jet Propulsion Laboratory, # 958208, as a part of the TOPEX-Poseidon investigation.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1993

Description: This thesis examines the use of a single, omnidirectional hydrophone as a receiving sensor to passively localize an acoustic beacon. The localization problem is presented as a constrained, nonlinear parameter estimation problem, and Lagrange multipliers are introduced to solve for the maximum likelihood estimate of the acoustic beacon's position. An iterative algorithm is developed using range difference measurements to solve for the maximum likelihood estimate of a stationary acoustic beacon's position. This algorithm is then _extended to include linear, constant velocity motion of the acoustic beacon. Finally, design specifications for a receiver to implement the maximum likelihood estimation algorithms are developed. To test the maximum likelihood estimate algorithms, Monte Carlo simulations are conducted. Results from six representative scenarios are presented. Test results show that as the number of range differences used increases, or the distance that the observer travels between received beacon signals increases, the accuracy of the estimated position improves. Also, tests show that accuracy of the estimated beacon position is directly related to the accuracy in which the observer's position is measured. To test the receiver's design specifications, a prototype receiver is built using commonly available components. It is then shown that the prototype receiver meets or exceeds the design specifications.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and Woods Hole Oceanographic Institution February 1994

Description: This thesis presents the design and simulation of an acoustic listen-only navigation system for use in a time-varying high-multipath environment. The system uses a ray tracing model of sound propagation in a horizontally stratified ocean to associate each multipath arrival with a particular ray path through the ocean. This ray path identification allows an inversion for both vehicle position and sound speed profile parameters. Tracking changes in the sound speed profile allows the vehicle to maintain navigation accuracy in a time-varying acoustic environment. Simulation results for typical Arctic conditions indicate the potential for accurate navigation from acoustic beacons at ranges up to tens of kilometers.

Description: I also wish to acknowledge the support received from a National Science Foundation Graduate Fellowship under Grant #RCD-9154652 and from the Office of Naval Research Arctic Programs Office, which made possible this work.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and Woods Hole Oceanographic Institution February 1994

Description: In this work we study motion of a baroclinic upper-ocean eddy over a large-scale topography which simulates a continental slope. We use a quasigeostrophic f-plane approximation with continuous stratification. To study this problem we develop a new numerical technique which we call "semi-lagrangian contour dynamics". This technique resembles the traditional 2-D contour dynamics method but differs significantly from it in the numerical algorithm. In addition to "Lagrangian" moving contours it includes an underlying "Eulerian" regular grid to which vorticity or density fields are interpolated. To study topographic interactions in a continuously stratified model we use density contours at the bottom in a similar manner as vorticity contours are used in the standard contour dynamics. For the case of a localized upper-ocean vortex moving over a sloping bottom the problem becomes computationally 2-dimensional (we need to follow only bottom density contours and the position of the vortex itself) although the physical domain is still 3-dimensional. Results of the numerical model lndicate importance of baroclinic effects in the vortex-topography interaction. After the initial surge of topographic Rossby waves a vortex moves almost steadily due to the interaction with a bottom density anomaly which is created and supported by a vortex itself. This anomaly is equivalent to a region of opposite-signed vorticity with a total circulation exactly compensating that of a vortex. This results in a vertically aligned dipolar structure with the total barotropic component equal to zero. Analytical considerations explaining this effect are presented and formulated in a more general statement which resembles but does not coincide with the "zero angular momentum theorem" of Flierl, Stern and Whitehead, 1983. In such steady translation the centroid of a bottom density anomaly is displaced horizon tally from the center of an upper-ocean vortex so the whole system moves due to this misalignment, which is known as a "he tonic mechanism". Cyclonic vortices go generally upslope, and anticyclones - in a downslope direction. The along-slope component of their motion depends upon the strength of a vortex, curvature of the bottom slope and background flows. When surrounded by a bowl-shaped topography anticyclonic vortices tend to stay near the deepest center of a basin, even resisting ambient flows which advect them outward. Application of this results to various oceanic examples (particularly to the "Shikmona eddy" in the Eastern Meditenanian) is discussed. Our results show that the behavior of a vortex over a sloping bottom differs significantly from its motion on the planetary beta-plane (but with a flat bottom). To explain this difference we introduce the concept of a "wave-breaking regime" relevant for the case of a planetary beta-effect, and a "wave-gliding regime" which characterizes the interaction of an eddy with a topographic slope.

Description: This work was supported by the NSF grant #OCE 90-12821.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and Woods Hole Oceanographic Institution February 1994

Description: Ambient noise in the sea has been observed for over 100 years. Previous studies conclude that the primary source of microseisms is nonlinear interaction of surface gravity waves at the sea surface. Though this source relationship is generally accepted, the actual processes by which the wave generated acoustic noise in the water column couples and propagates to and along the sea floor are not well understood. In this thesis, the sources and propagation of sea floor and sub-sea floor microseismic noise between 0.2 and 10 Hz are investigated. This thesis involves a combination of theoretical, observational and numerical analysis to probe the nature of the microseismic field in the Blake Bahama Basin. Surface waves are the primary mechanism for noise propagation in the crust and fall into two separate groups depending on the relative wavelength/water depth ratio. Asymptotic analysis of the Sommerfeld integral in the complex ray parameter plane shows results that agree with previous findings by Strick (1959) and reveal two fundamental interface wave modes for short wavelength noise propagation in the crust: the Stoneley and pseudo-Rayleigh wave. For ocean sediments, where the shear wave velocity is less than the acoustic wave velocity of water, only the Stoneley interface wave can exist. For well consolidated sediments and basalt, the shear velocity exceeds the acoustic wave velocity of water and the pseudo-Rayleigh wave can also exist. Both interface waves propagate with retrograde elliptic motion at the sea floor and attenuate with depth into the crust, however the pseudo-Rayleigh wave travels along the interface with dispersion and attenuation and "leaks" energy into the water column for a half-space ocean over elastic crust model. For finite depth ocean models, the pseudo-Rayleigh wave is no longer leaky and approaches the Rayleigh wave velocity of the crust. The analysis shows that longer wavelength noise propagates as Rayleigh and Stoneley modes of the ocean+crust waveguide. These long wavelength modes are the fundamental mechanism for long range noise propagation. During the Low Frequency Acoustic Seismic Experiment (LFASE) a four-node, 12- channel borehole array (SEABASS) was deployed in the Blake Bahama Basin off the coast of eastern Florida (DSDP Hole 534B). This experiment is unique and is the first use of a borehole array to measure microseismic noise below the sea floor. Ambient background noise from a one week period is compared between an Ocean Bottom Seismometer (OBS) and SEABASS at sub-bottom depths of 10, 40, 70 and 100 meters below the sea floor. The 0.3 H z microseism peak is found to be nearly invariant with depth and has a power level of 65 and 75 dB rel 1 (nm/ s2)2)/ H z for the vertical and horizontal components respectively. At 100 m depth, the mean microseismic noise levels above 0.7 Hz are 10 dB and 15-20 dB quieter for the vertical and horizontal components respectively. Most of this attenuation occurs in the upper 10 m above 1.0 Hz, however higher modes in the spectra show narrow bandwidth variability in the noise field that is not monotonic with depth. Dispersion calculations show normal mode Stoneley waves below 0.7 Hz and evidence of higher modes above 0.8 Hz. A strong correlation between noise levels in the borehole and local sea state conditions is observed along with clear observation of the nonlinear frequency doubling effect between ocean surface waves and microseisms. Particle motion analysis further verifies that noise propagates through the array as Rayleigh/Stoneley waves. Polarization direction indicates at least two sources; distant westerly swell during quiescent times and local surface waves due to a passing storm. Above 1.0 Hz the LFASE data shows little coherence and displays random polarization. Because of this, we believe scattered energy is a significant component of the noise field in the Blake Bahama Basin. A fully 3-D finite difference algorithm is used to model both surface and volume heterogeneities in the ocean crust. Numerical modeling of wave propagation for hard and soft bottom environments shows that heterogeneities on the order of a seismic wavelength radiate energy into the water column and convert acoustic waves in the water into small wavelength Stoneley waves observed at the borehole. Sea floor roughness is the most important elastic scattering feature of the ocean crust. Comparisons of 2D and 3D rough sea floor models show that out-of-plane effects necessitate the use of 3D methods. The out-of-plane energy that is present in the LFASE data comes from either heterogeneities in the source field (i.e. mixed gravity wave directions) or, equally likely, scattering of the source field from surface or volume heterogeneities in the sea floor.

Description: This research was supported by Office of Naval Research grants N00014-89-C-0018, N00014-89-J-1012, N00014-90-C-0098, N00014-90-J-1493 and N00014-93-1-1352.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and Woods Hole Oceanographic Institution October 1993

Description: Hydrographic and expendable current profiler (XCP) data taken during the Gulf of Cadiz Expedition in September 1988 are analyzed to diagnose the mixing and dynamics of the Mediterranean outflow. The overall structure of the outflow is consistent with that described in the historical literature (Heezen and Johnson, 1969). This data shows that the overflow transport doubles from .85 Sv to 1.9 Sv, and that the velocity weighted salinity decreases from 37.8 pss to 36.7 pss in the first 60 km of the path. The core salinity of the neutrally buoyant outflow near Cape St. Vincent is 36.6 pss, which indicates that most of the mixing has taken place close to the Strait in the initial descent of the outflow. Cross stream variations in the overflow T/S properties increase as the flow spreads from 10 km to 90 km wide. The outflow begins with less than a 0.5°C across-stream variation in temperature in the Strait with the saltiest, coldest water to the south and slightly fresher and warmer outflow to the north. As the outflow spreads, the northern near-shelf flow follows a path higher in the water column and mixes with warmer North Atlantic water than does the deeper offshore flow. Within the first 100 km, the cross stream variation in temperature on an isopycnal becomes more than a 2°C. The flow eventually settles along two preferred isopycnals: 27.5 and 27.8 (Zenk 1975b). The spreading of the flow contains both a barotropic and baroclinic character. The average change in angle above and below the maximum velocity of the outflow is 8°while at the edges of the flow the average direction of the outflow diverges by as much as 50°. Gradient Richardson numbers less than 1/4 are found in the interface (up to 50 m thick) between westward flowing Mediterranean water and eastward flowing North Atlantic water, even though there is a strong stabilizing stratification present. Bulk Froude numbers greater than 1 are found near the Strait coincident with the vigorous mixing noted above. Lower bulk Froude numbers were observed in regions where less entrainment was taking place. The momentum balances are diagnosed using hydrographic and XCP data. Evaluation of the cross stream momentum balance shows the importance of advection as the flow makes a 90 degree inertial turn upon entering the Gulf of Cadiz. A form of the Bernoulli function can be evaluated to infer the total stress (entrainment and bottom drag) acting on the outflow. This stress is as large as 5 Pa within 20 km of the Strait, while further downstream the stress decreases to about 1/2 Pa. The entrainment stress estimated from the property fluxes reaches a maximum of about 0.8 Pa near section C, indicating that bottom stress is dominant. Near the Strait, advection, bottom drag and the Coriolis force are all critical to the dynamics of the outflow. Further downstream, the outflow becomes a damped geostrophic current. A simple geostrophic adjustment model is used to show that in the absence of frictional stresses, the outflow would very quickly become geostrophically balanced and descend only about 10 m down the continental slope. Thus, friction is critical for the outflow to cross isobaths. A simple numerical model that uses a Froude number dependent entrainment and a quadratic bottom friction law is used to simulate the outflow (Price and Baringer, 1993). Some of the properties of the outflow including localized entrainment, large stresses and high Rossby number of the flow (initially as high as 0.6), are simulated rather well, though the model overestimates the magnitude of the outflow current. We suspect that this is a consequence of assuming a passive ocean. Two different methods for specifying the broadening of the flow are compared: one using the highly parameterized concept of Ekman spreading, the other using the conservation of potential vorticity. The potential vorticity broadening more accurately reproduces the observed width of the flow near Cape St. Vincent where the width varies inversely with the bottom slope. However, both methods produce essentially the same equilibrium temperature, salinity and transport of the outflow which is a testament to the robustness of the model solution. with the formation process of NADW.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1996

Description: During January-March, Scotian Shelf water has been observed to flow episodically from the southwestern Scotian Shelf directly across the Northeast Channel to Georges Bank. The possible factors that allow Scotian Shelf water to break the topographic constraint presented by the Northeast Channel and flow directly to Georges Bank are considered. As a simple analog to the flow over the southwestern Scotian Shelf near the Northeast Channel, the adjustment of a barotropic current near a shelf-break to a sharp bend in the shelf topography is studied numerically. For parameters within the oceanographic range, the adjustment to the bend is smooth and steady with no eddies shed at the corner. The vorticity dynamics allow a balance between the vortex stretching in the flow and the curvature in the flow. This is possible since the bend is a right-hand one facing downstream, a similar balance not being possible for a left-hand bend, in which case eddy formation is likely. A simple model of this balance clarifies the vorticity dynamics and provides the scaling rc = √eL/0.765 for any streamline in the flow, where rc is the radius of curvature at the corner, E = u0/fL and L = h0/b, where uo is the initial speed, f the coriolis parameter, h0 the initial depth and b the bottom slope. These results show that other factors such as stratification, wind stress, and time-dependent inflow must play a role in any flow across the Northeast Channel.

Description: I am very grateful to the US-GLOBEC program for providing the funding for this study (N.S.F. grant OCE-9313671).

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1995

Description: Geosat altimeter data and numerical model output are used to examine the circulation and dynamics of the Antarctic Circumpolar Current (ACC). The mean sea surface height across the ACC has been reconstructed from height variability measured by the Geosat altimeter, without assuming prior knowledge of the geoid. For this study, an automated technique has been developed to estimate mean sea surface height for each satellite ground track using a meandering Gaussian jet model, and errors have been estimated using Monte Carlo simulation. The results are objectively mapped to produce a picture of the mean Subantarctic and Polar Fronts, which together comprise the major components of the ACC. The locations of the fronts are consistent with in situ observations and indicate that the fronts are substantially steered by bathymetry. The jets have an average Gaussian width of about 44 km in the meridional direction and meander about 75 km to either side of their mean locations. The width of the fronts is proportional to 1/f, indicating that with constant stratification, the width is proportional to the baroclinic. Rossby radius. The average height difference across the Subantarctic Front (SAF) is 0.7 m and across the Polar Front (PF) 0.6 m. The mean widths of the fronts are correlated with the size of the baroclinic Rossby radius. The meandering jet model explains between 40% and 70% of the height variance along the jet axes. Bathymetric constrictions are associated with increased eddy variability, a smaller percentage of which may be explained by the meandering of the ACC fronts, indicating that propagating eddies and rings may be spawned at topographic features. Detailed examination of spatial and temporal variability in the altimeter data indicates a spatial decorrelation scale of 85 km and a temporal e-folding scale of 34 days. The sea surface height variability is objectively mapped using these scales to define autocovariance functions. The resulting maps indicate substantial evidence of mesoscale eddy activity. Over 17-day time intervals, meanders of the PF and SAF appear to elongate, break off as rings, and propagate. Statistical analysis of ACC variability from altimeter data is conducted using empirical orthogonal functions (EOFs ). The first mode EOF describes 16% of the variance in total sea surface height across the ACC; reducing the domain into basin scales does not significantly increase the variance represented by the first EOF, suggesting that the scales of motion are relatively short, and may be determined by local instability mechanisms rather than larger basin scale processes. Likewise, frequency domain EOFs indicate no statistically significant traveling wave modes. The momentum balance of the ACC has been investigated using both output from a high resolution primitive equation model and sea surface height measurements from the Geosat altimeter. In the Semtner-Chervin general circulation model, run with approximately quarter-degree resolution and time varying ECMWF winds, topographic form stress is the dominant process balancing the surface wind forcing. Detailed examination of form stress in the model indicates that it is due to three large topographic obstructions located at Kerguelen Island, Campbell Plateau, and Drake Passage. In order to reduce the effects of standing eddies, the model momentum balance is considered in stream coordinates; vertically integrated through the entire water column, topographic form drag is the dominant balance for wind stress. However, at mid-depth the cross-stream momentum transfer is dominated by horizontal biharmonic friction. In the upper ocean, horizontal friction, mean momentum flux divergence, transient momentum flux divergence, and mean vertical flux divergence all contribute significantly to the momentum balance. Although the relative importance of individual terms in the momentum balance does not vary substantially along streamlines, elevated levels of eddy kinetic energy are associated with the three major topographic features. In contrast, altimeter data show elevated energy levels at many more topographic features of intermediate scales, suggesting that smaller topographic effects are better able to communicate with the surface in the real ocean than in the model. Transient Reynolds stress terms play a small role in the the overall momentum balance; nonetheless, altimeter and model measurements closely agree, and suggest that transient eddies tend to accelerate the mean flow, except in the region between the major fronts which comprise the ACC. Potential vorticity is considered in the model output along Montgomery streamfunction. Even at about 1000 m depth, it varies in response to wind forcing, largely as a result of changes in vertical stratification, indicating that forcing and dissipation do not locally balance in the Southern Ocean. In order to compare model and altimeter potential vorticity estimates, two different proxies for potential vorticity on surface streamlines are considered. Both proxies show very similar results for model and altimeter, suggesting that differences in surface streamlines estimated by the altimeter and the model are not significant in explaining the Southern Ocean flow. The proxies are both roughly conserved along surface height contours but undergo substantial jumps near topographic features. However, they cannot capture stratification changes which may be critically important to the overall potential vorticity balance.

Description: Funding for this research was provided by an Office of Naval Research graduate student fellowship and National Aeronautics and Space Administration contract NAGW-1666.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1991

Description: A set of four hydrographic sections through the Brazil Current are analyzed to identify downstream changes in the Brazil Current. The data, from the Thomas Washington Marathon Cruise, Leg 9, are at 27, 31, 34 and 36°S. The region they span details the change of the current from a relatively small near surface feature to a large, deep current. While the Brazil Current does not appear to develop transports as large as those found in the Gulf Stream, the calculated transports greatly exceed previous estimates. At 27°S the current extends down to approximately 700 m and transports 12 Sv southward; this value is consistent with previous estimates farther north. Downstream, surface layer transport increases, the current deepens, and the transport reaches a maximum of 80 Sv at 36°S. Part of the growth comes from the tight recirculation found just offshore of the Brazil Current. The recirculation strengthens and deepens to the south, with a minimum transport of 4 Sv north at 27°S and a maximum of 33 Sv at 36°S. The change in the current is also reflected in its shear profiles. At 27°S Brazil Current shear is found only in the upper portion of the water column, over the continental slope. Downstream, the current moves off the slope into deeper water and develops top-to-bottom, monotonic shear. To obtain velocity from the shear profiles, zero velocity surfaces are chosen based on conservative use of tracer information. A simple basin-wide model is used at 31°S to tie limits on the size of the Brazil Current and recirculation to various limits on layer-to-layer exchanges south of the section. The multi-layer model - based on changes in depth of several isotherms is used to extend the interpretation of the current beyond that of an isolated ocean feature. The model is required to conserve mass in each layer, either by applying barotropic transports or by allowing layer-to-layer exchanges south of the section. Solutions are deemed acceptable if the sense, or direction, of the various layer-to-layer conversions are consistent with accepted ideas of water mass formation. Initially, a two layer model is employed. Governed by the conservation of mass in each layer, the two layer model has only one constraint on the resulting solutions: a conversion of cold-to-warm water in the south (or the surface layer flowing north and the deep layer flowing south). Such a meridional flow pattern is consistent with the equatorward heat flux in the South Atlantic. The single constraint, however, is not strong enough to limit the solution region in any significant way. The final model presented has four layers, and acceptable solutions have the net transports of the surface layer and the bottom water northward and form intermediate water from North Atlantic Deep Water in the south. The resulting solution set has a fairly small range of transports for the Brazil Current, with surface layer transports between 20 and 35 Sv; this range is consistent with the value calculated from hydrographic data at 31°S. Given the complex interleavings of the South Atlantic water masses, the four layer model performs remarkably well. Finally, total potential vorticity is calculated from the hydrographic sections. Contrary to what one might expect, the reference level choice is not a significant problem: where currents are large, most of the signal in relative potential vorticity comes from the measured shear, and where currents are small, the relative potential vorticity is not significant compared to the planetary vorticity. Unfortunately, the process of taking two horizontal derivatives of the density field results in a jittery relative potential vorticity signal. As a result, a clear potential vorticity profile could not be constructed for the current. This variablitiy may be real -the ocean is frequently much noisier than one imagines. It may also be possible, though, to smooth the data sufficiently so that a cleaner picture emerges. Despite the problems involved in obtaining a quantitative profile of the potential vorticity, qualitative changes are useful in detecting different flow regimes. By comparing the downstream changes in total and planetary potential vorticity, one can deduce frictional and inertial regimes in the different layers. The presence of a frictional regime at the inshore edge suggests that care should be taken in assuming that potential vorticity is conserved in western boundary currents. In addition the potential vorticity sections trace a pattern of the recirculation feeding into the Brazil Current in the upper layers; other tracers did not provide a clear image. The final picture which emerges is not of a small, surface-trapped Brazil Current; rather, it is that of a classic western boundary current, increasing in strength and depth before turning east into the interior ocean.

Description: Financial support for the data collection and initial analysis was provided through the Office of Naval Research South Atlantic Accelerated Research proposal under contract N00014-82-C-0019. Continued analysis was supported by the National Science Foundation under grant OCE86-14486.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution August 1991

Description: Using a semi-geostrophic, reduced gravity thin jet model, we analytically study the evolution of initial meanders into pinched-off rings. The model used is similar to the path equation developed by Flierl and Robinson {1984) for vertically coherent meanders. However, in the present model, the meanders are baroclinic, and a stretching term arises due to the motion of the interface. It can be shown that the equation governing the time-dependent meander of this jet (Pratt, 1988) can be transformed into the Modified Korteweg- deVries (MKdV) equation in intrinsic coordinates. The MKdV equation admits two types of solitary wave solutions, loop solitons and breathers. The breathers are permanent meanders which propagate on the path , and some are able to form rings. Using the inverse scattering transform , we can predict breather and ring formation for simple initial meanders. The inverse scattering t ransform is applied to S and Ω shaped meanders with piecewise constant and continuous curvature. S shaped meanders, or steps, must be multi-valued to form breathers, and must have very steep angles in order to form rings. Due to integral constraints, Ω shaped meanders, or lobes, are unable to pinch together to form rings unless they are wide enough so that the two side flanks of the lobe act as two independent steps. The numerical solutions indicate that the breathers predicted by the inverse scattering is a very good approximation to the full solution.

Description: This work has been supported by NSF contract number OCE 89-16446 and ONR contract number N00014-S9-J-1182.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution March 1992

Description: Rotating baroclinic and barotropic boundary currents flowing around a corner in the laboratory were studied in order to discover the circumstances under which eddies were produced at the corner. Such flows are reminiscent of oceanic coastal flows around capes. When the baroclinic currents, which consisted of surface flows bounded by a density front, encountered a sharp corner, immediately downstream of the corner an anticyclone grew in the surface layer for an angle of greater than 40 degrees. Varying the initial condition of the flow or the depth of the lower layer did not noticeably affect the gyre's properties except for its growth speed, which was greater when the lower layer was shallower. The barotropic currents were pumped along a sloping bottom, and also formed anticyclonic gyres which quickly attained an approximately steady state. For a given topography, the size of the gyre was proportional to the inertial radius u/f. Volume flux calculations based on the surface velocity revealed vertical shear which increased with gyre size. Hydraulic models were also applied to flow around gently curving topography to determine the critical separation curvature as a function of upstream parameters.

Description: This work was supported by the National Science Foundation grant OCE 89-15408.

Description: Submitted in partial fulfillment of the requirements for the degree of Ocean Engineer at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1990

Description: This paper develops a beam pattern design procedure for general multidimensional irregular sonar arrays that incorporates the not well understood effects of array geometry into the design process. The procedure is implemented by generating a "penalty function" in a spectral covariance function form. Processing the penalty function causes beam pattern high sidelobes to be penalized and the main lobe to be emphasized. This is accomplished by forming the penalty function in terms of an isotropic noise field of specified strength modified with a finite sector of low coherent energy and stabilized with incoherent sensor noise. By inputting the penalty function into a minimum variance beamformer, the beam pattern and aperture weights are calculated based on the given array geometry. The beamformer used is Capon's Maximum Likelihood Method. The array used to test the procedure is located on a sixty degree sector of a cylindrical surface. The procedure is implemented by two different methods, each with some desirable characteristics. One method suppresses sidelobes directly by the placement of nulls. The other method suppresses sidelobes indirectly by the enhancement of the main lobe with anti-nulls. Both methods are evaluated in terms of a sensitivity factor which constrains the maximum white noise array gain. Results show that both methods result in sidelobe levels that range from 20 to 35 dB lower compared to a conventional beam pattern with uniform aperture weighting and that the design procedure is applicable to beam patterns steered to both true broadside and to off-broadside directions.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1991

Description: This thesis examines the effect of mean large-scale currents on the vertical structure of the upper ocean during two recent observational programs: the Long Term Upper Ocean Study (LOTUS) and the TROPIC HEAT experiments. The LOTUS experiment took place in the northwest Atlantic Ocean, a mid-latitude region away from strong mean currents, and extended over one entire seasonal cycle. The TROPIC HEAT experiments took place in the central equatorial Pacific Ocean during two 12-day periods in 1984 and 1987, at opposite extremes of the seasonal cycle. We use observations from these field experiments as well as one-dimensional numerical models of the upper ocean to analyze the dynamics of the vertical structure of the upper ocean at the equator and in mid-latitudes. Due to the different nature of the observations, we focus on the long term mean structure of the upper ocean in the LOTUS observations (Chapters 2 and 3), and on the diurnal cycle in the equatorial upper ocean in our analysis of the TROPIC HEAT observations (Chapters 4 and 5). In the LOTUS observations, we find that the observed current is coherent with the wind over low frequencies (greater than an inertial period). Using a wind-relative averaging method we find good agreement with Ekman transport throughout the first summer and winter of the LOTUS experiment, with the exception of a downwind component in the wintertime. The mean current spiral is flat compared to the classic Ekman spiral, in that it rotates less with depth than does the Ekman spiral. The mean current has an e-folding depth scale of 12m in the summer and 25 min the winter. Diurnal cycling is the dominant variability in the summer and determines the vertical structure of the spiral. In the winter, diurnal cycling is almost non-existent due to greatly reduced solar insolation. There is a persistent downwind shear in the upper 15 m during the winter which may be partially due to a bias induced by surface wave motion but which is also consistent with a logarithmic boundary layer. The Price et al. (1986) model is reasonably successful in simulating the current structure during the summer, capturing both the mean and the diurnal variation. The model is less successful in the winter, though it does capture the overall depth scale of the current spiral. In our analysis of the TROPIC HEAT observations, we extend the Price et al. (1986) model to the equatorial upper ocean. The model is initialized with the stratification and shear of the Equatorial Undercurrent (EUC), and is driven with heating and wind stress. A surface mixed layer is determined by bulk stability requirements, and a transition layer below the mixed layer is simulated by requiring that the gradient Richardson number be no less than 1/4. A principal result is that the nighttime phase of the diurnal cycle is strongly affected by the EUC, resulting in deep mixing and large dissipation at night consistent with observations of the equatorial upper ocean during TROPIC HEAT. Other features of the equatorial circulation (upwelling and the zonal pressure gradient) are of little direct importance to the diurnal cycle. The daytime (heating) phase of the simulated diurnal cycle is unaffected by equatorial circulation and is very similar to its mid-latitude counterpart. Solar heating produces a stably stratified surface layer roughly 10 m thick within which there is little, 0(3 x 10-8 W kg-1), turbulent dissipation. The diurnal stratification, though small compared to the EUC, is sufficient to insulate the EUC from wind stress during the day. For the typical range of conditions at the equator, diurnal warming of the sea surface is 0.2-0.5°C, and the diurnal variation of surface current (diurnal jet) is 0.1-0.2 m s-1, consistent with observations. The nighttime (cooling) phase of the simulated diurnal cycle is quite different from that seen at mid-latitudes. As cooling removes the warm, stable surface layer, the wind stress can work directly against the shear of the EUC. This produces a transition layer that can reach to 80 m depth, or nearly to the core of the EUC. Within this layer the turbulent dissipation is quite large, 0(2 x 10-7 W kg-1). Thus, the simulated dissipation has a diurnal range of more than a factor of five, as observed in the 1984 TROPIC HEAT experiment, though the diurnal cycle of stratification and current are fairly modest. Dissipation estimated from the model is due to wind working directly against EUC, and is similar to observed values of dissipation in both magnitude and depth range. Overall dissipation values in the model are set by the strength of the wind stress rather than the structure of the EUC, and rise approximately like u*3 for a given Undercurrent. This suggests that the lower values of dissipation observed in the 1987 TROPIC HEAT experiment were due to the lower wind stress values rather than the relatively weak Undercurrent. The main findings of this thesis are: 1) When the diurnal cycle in solar heating is strong, it determines the local vertical structure of the upper ocean (in both the LOTUS and TROPIC HEAT observations). The Price et al. (1986) model and its extension to the equator simulate the upper ocean fairly well when the diurnal cycle is strong. Under these conditions it is necessary to make measurements very near the surface ( 〈 10 m depth) to fully resolve the wind-driven flow. 2) When surface waves are strong, surface-moored measurements of current may have a significant wave bias. To accurately estimate this bias, simultaneous measurements of current, current meter motion, and surface waves are needed. 3) Mean currents strongly amplify the nighttime phase of the diurnal cycle in the equatorial upper ocean, and therefore alter the mean structure of the equatorial upper ocean.

Description: The Office of Naval Research supported this work under contract N00014-89-J-1053.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1992

Description: This study concerns the barotropic interactions between a mesoscale eddy and a straight monotonic bottom topography. Through simple to relatively complicated modeling effort, some of the fundamental properties of the interaction are investigated. In chapter two, the fundamental aspects of the interaction are examined using a simple contour dynamics model. With the simplest model configuration of an ideal vortex and a step topography, the basic dynamical features of the observed oceanic eddy-topography interaction are qualitatively reproduced. The results consist of eddy-induced cross-topography exchange, formation of topographic eddies, eddy propagation and generation of topographic waves. In chapter three, a more complicated primitive equation model is used to investigate a mesoscale eddy interacting with an exponential continental shelf/slope topography on both f and β-planes. The f-plane model recasts the important features of chapter two. The roles of the eddy size and strength and the geometry of topography are studied. It is seen that the multiple anticyclonic eddy-slope interactions strongly affect the total cross-slope volume transport and the evolution of both the original anticyclone and the topographic eddy. Since a cyclone is trapped at the slope and eventually moves on to the slope, it is most effective in causing perturbation on the shelf and slope. The responses on the shelf and slope are mainly wavelike with dispersion relation obeying that of the free shelf-trapped wave modes. On the β-plane, the problem of an eddy colliding onto a continental shelf/slope from a distance with straight or oblique incident angles is investigated. It is found that the straight eddy incident is more effective in achieving large onslope eddy penetration distance than the oblique eddy incident. The formation of a dipole-like eddy pair consisting of the original anticyclone and the topographic cyclone acts to suppress the eddy decay due to long Rossby wave radiation. A weak along-slope current near the edge of the slope is found, which is part of a outer slope circulation cell originated from the Rossby wave wake trailing the propagating eddy. Model-observation comparisons in_chapter four show favorable qualitative agreement of the model results with some of the observed events in the eastern U.S. continental margins and in the Gulf of Mexico. The model results give dynamical interpretations to some observed features of the oceanic eddy-topography interactions and provide enlightening insight into the problem.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2012

Description: Signal detection and synchronization in the time varying ocean environment is a difficult endeavor. The current common methods include using a linear frequency modulated chirped pulse or maximal length sequence as a detection pulse, then match filtering to that signal. In higher signal to noise ratio (SNR) environments (~0 dB and higher) this has been a suitable solution. As the SNR drops lower however, this solution no longer provides an acceptable probability of detection for a given tolerable probability of false alarm. The issue derives from the inherent coherence issues in the ocean environment which limit the useful matched filter length. This thesis proposes an alternative method of detection based on a recursive least squares linearly adaptive equalizer which we term the Adaptive Linear Equalizer Detector (ALED). This detectors performance has demonstrated reliable probability of detection with minimal interfering false alarms with SNR as low as -20 dB. Additionally this thesis puts forth a computationally feasible method for implementing the detector.

Description: Support from the Office of Naval Research (through ONR grant #N00014-07-10738 and #N00014-11-10426).

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1990

Description: The contribution of tropical instability waves to the momentum and energy balances of the Pacific Equatorial Undercurrent is investigated using velocity and temperature time series from the three-dimensional Equatorial Pacific Ocean Climate Study mooring array at 110°W. Tropical instability waves are an energetic band of variability typically with periods between 14 and 36 days which are thought to be generated by instability of the equatorial currents. They are frequently observed as meanders of the equatorial front in satellite sea surface temperature maps. Here, they are observed as large oscillations in the meridional velocity records at l10°W with an energy peak at 21 days. Westward phase propagation is observed in this band with a phase speed of -0.9 (±0.3) m s-1 and a wavelength of 1660 km. Upward phase propagation is observed which is consistent with downward energy propagation. The observed propagation characteristics are compared with those of the mixed Rossby-gravity wave. The variability in this band produces large northward fluxes of eastward momentum and southward fluxes of temperature which affect the dynamics of the mean Undercurrent through the Reynolds stress divergence, and the Eliassen-Palm flux divergence. The waves produce a northward flux of eastward momentum, uv, which is largest at the northern mooring in the upper part of the array. The meridional divergence of eastward momentum, -δ(uv)/δy, decelerates the Undercurrent core down to 150 m. This implies a coupling between the Undercurrent and the South Equatorial Current with the eastward momentum of the Undercurrent transferred to the westward flowing South Equatorial Current. To estimate the vertical momentum flux divergence, the vertical eddy flux of eastward momentum, uw, is inferred using the eddy temperature equation. The vertical eddy momentum flux is positive and largest at the core of the Undercurrent, implying an acceleration of the eastward flow above the core and a deceleration below. The Eliassen-Palm flux divergence is small above the core of the Undercurrent at 75 m, but below the core, is sufficient to balance the deeply penetrating eastward pressure gradient force. The instability waves are important to the energetics of the mean Undercurrent. An exchange of kinetic energy from the mean Undercurrent to the waves through shear production is estimated. A local exchange is suggested since the rate at which the mean Undercurrent loses kinetic energy through instability is comparable to the rate at which the waves gain energy through shear production. The conversion from mean to eddy potential energy is an order of magnitude smaller with the waves gaining potential energy through conversion of mean available potential energy. The observations of upward phase propagation and downward Eliassen-Palm flux suggest that the waves propagate energy downward into the deep ocean. The energetics and momentum balance of the mean Undercurrent is investigated further by analyzing the downstream change in the Bernoulli function on the equator along isentropes or potential density surfaces using mean hydrographic sections at 150°W and 110°W. A downstream decrease in the Bernoulli function is observed which is due to a decrease in the Acceleration Potential since the mean kinetic energy of the Undercurrent changes little from 150°W to 110°W. The lateral divergence of eddy momentum fluxes calculated on isotherms is sufficient to balance the observed decrease in the Acceleration Potential. The downstream decrease in the Acceleration potential has further implications for the mean energetics since this "downhill" flow releases mean available potential energy stored in the east-west sloping thermocline. The rate at which the Undercurrent releases available potential energy, is shown to be comparable to the rate at which the mean flow loses kinetic energy by interaction with the waves, with the waves gaining kinetic energy in the process. Thus, it is hypothesized that in the eastern Pacific this downstream release of available potential energy is ultimately converted into a downstream increase in the kinetic energy of the waves rather than the kinetic energy of the mean flow as occurs in the western Pacific. To maintain an equilibrium, the waves radiate energy into the deep ocean as is suggested by the upward phase propagation and the downward Eliassen-Palm flux.

Description: Financial support of the National Science Foundation under contracts OCE 82-14955 and OCE 85-19551, for participation in Tropic Heat, and OCE 85-04125.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1999

Description: This thesis studies the problems of generation and maintenance of recirculations by Gulf Stream instabilities. Observations show that the horizontal structure of the jet and its recirculations suffer significant changes in time. Here, the role of internal dynamics of the jet is isolated as one of the possible sources of such variability, and the differences between barotropic and baroclinic instabilities are investigated. The problem of recirculation development is considered in a framework of a free spin down of the 2-layer and the 1-layer, zonally symmetric, quasi-geostrophic jets. Linear stability analysis shows that in strongly baroclinic basic flows, eddies are capable of driving recirculations in the lower layer through the residual meridional circulation. In strongly barotropic jets, the linearly most unstable wave simply diffuses the jet. Nonlinear stability analysis indicates that recirculations are robust features of the 2-layer model. The strength of recirculations is a function of the model’s parameters. It increases with a decrease in the value of the nondimensional /3 due to potential vorticity homogenization constrained by enstrophy conservation. The recirculation strength is a non-monotonic function of the baroclinic velocity parameter; it is the strongest for strongly baroclinic basic flows, weakest for flows with intermediate baroclinic structure and of medium strength for strongly barotropic basic flows. Such non-monotonic behavior is the result of two different processes responsible for the recirculation development: linear eddy-mean flow interactions for strongly baroclinic basic flows and strongly nonlinear eddy-eddy and eddy-mean flow interaction for strongly barotropic flows. In the case of the reduced-gravity model, recirculations develop only for infinite deformation raduis. Basic flows with finite deformation radius are only weakly supercritical and therefore produced negligible recirculations after equilibration. The problem of maintenance of the recirculations is coupled to the questions of existence of low frequency variability and of multiple dynamical regimes of a system consisting of a quasi-geostrophic jet and its recirculations. The problem is studied in a framework of a 2-layer or a reduced-gravity colliding jets model which has no windforcing. Instead, it is forced by inflows and outflows through the open boundaries. Oniy the western boundary of the domain is closed, and a free slip boundary condition is used there. The results of the numerical experiments show that when oniy the mechanism of barotropic instability is present, the model has two energy states for a wide range of interfacial friction coefficients. The high energy state is characterized by well-developed recirculations and displays strong variability associated with either large recirculating gyres and a weak eddy field or small recirculations and a strong eddy field. The iow energy state is characterized by large meridional excursions in the separation point and large amplitude, westward propagating meanders that produce strong rings after interacting with the western wall. For physically relevant bottom friction values, the presence of baroclinic in stability in the recirculation regions of the 2-layer model allows for a unique dynamical regime characterized by well-developed recirculations in both layers. The low-frequency variability associated with the regime is weak and is related to meridional shifts in the position of the jet, to wrapping of the recirculations around each other, and to pulsations in their zonal extent. For strong bottom friction, the 2-layer model has only the mechanism of barotropic instability which reduces it to a 1 1/2-layer configuration; the model displays two dynamical regimes and strong low frequency variability in the upper layer, while the lower layer is strongly frictional.

Description: Financial support for this research was provided by NSF grant number OCE 9617848.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1999

Description: During July and August of 1996, a large acoustics/physical oceanography experiment was fielded in the Mid-Atlantic Bight, south of Nantucket Island, MA. Known as the Shelfbreak Front PRIMER Experiment, the study combined acoustic data from a moored array of sources and receivers with very high resolution physical oceanographic measurements. This thesis addresses two of the primary goals of the experiment, explaining the properties of acoustic propagation in the region, and tomographic inversion of the acoustic data. In addition, this thesis develops a new method for predicting acoustic coherence in such regions. Receptions from two 400 Hz tomography sources, transmitting from the continental slope onto the shelf, are analyzed. This data, along with forward propagation modeling utilizing SeaSoar thermohaline measurements, reveal that both the shelfbreak front and tidally-generated soliton packets produce stronger coupling between the acoustic waveguide modes than expected. Arrival time wander and signal spread show variability attributable to the presence of a shelf water meander, changes in frontal configuration, and variability in the soliton field. The highly-coupled nature of the acoustic mode propagation prevents detailed tomographic inversion. Instead, methods based on only the wander of the mode arrivals are used to estimate path-averaged temperatures and internal tide "strength". The modal phase structure function is introduced as a useful proxy for acoustic coherence, and is related via an integral transform to the environmental sound speed correlation function. Advantages of the method are its flexibility and division of the problem into independent contributions, such as from the water column and seabed.

Description: Office of Naval Research for providing funding for this thesis through AASERT Grant N00014-96-1-0918, and through ONR Grant N00014-98-1-0059.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1999

Description: In this thesis the analysis of natural ice events is carried out based on direct measurements of ice-borne seismo-acoustic waves generated by ice fracturing processes. A major reason for studying this phenomenon is that this acoustic emission is a significant contributor to Arctic ocean ambient noise. Also the Arctic contains rich mineral and oil resources and in order to design mining facilities able to withstand the harsh environmental conditions, we need to have a better understanding of the processes of sea ice mechanics. The data analyzed in this thesis were collected during the Sea Ice Mechanics Initiative SIMI’94 experiment which was carried out in the spring of 1994 in the Central Arctic. One of the contributions of this thesis was the determination of the polarization characteristics of elastic waves using multicomponent geophone data. Polarization methods are well known in seismology, but they have never been used for ice event data processing. In this work one of the polarization methods so called Motion Product Detector method has been successfully applied for localization of ice events and determination of polarization characteristics of elastic waves generated by fracturing events. This application demonstrates the feasibility of the polarization method for ice event data processing because it allows one to identify areas of high stress concentration and "hot spots" in ridge building process. The identification of source mechanisms is based on the radiation patterns of the events. This identification was carried out through the analysis of the seismo-acoustic emission of natural ice events in the ice sheet. Previous work on natural ice event identification was done indirectly by analyzing the acoustic energy radiated into the water through coupling from elastic energy in the ice sheet. After identification of the events, the estimation of the parameters of fault processes in Arctic ice is carried out. Stress drop, seismic moment and the type of ice fracture are determined using direct near-field measurements of seismo-acoustic signals generated by ice events. Estimated values of fracture parameters were in good agreement with previous work for marginal ice zone. During data processing the new phenomenon was discovered: "edge waves", which are waves propagating back and forth along a newly opened ice lead. These waves exhibit a quasi-periodic behavior suggesting some kind of stick-slip generation mechanism somewhere along the length of the lead. The propagation characteristics of these waves were determined using seismic wavenumber estimation techniques. In the low frequency limit the dispersion can be modeled approximately by an interaction at the lead edges of the lowest order, antisymmetric modes of the infinite plate.

Description: Support for this thesis was provided by Office of Naval Research.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2011

Description: Remote sensing and in situ observations are used to investigate the ocean response to the Tokar Wind Jet in the Red Sea. The wind jet blows down the pressure gradient through the Tokar Gap on the Sudanese coast, at about 18°N, during the summer monsoon season. It disturbs the prevailing along-sea (southeastward) winds with strong cross-sea (northeastward) winds that can last from days to weeks and reach amplitudes of 20-25 m/s. By comparing scatterometer winds with along-track and gridded sea level anomaly observations, it is shown that an intense dipolar eddy spins up in less than seven days in response to the wind jet. The eddy pair has a horizontal scale of 140 km. Maximum ocean surface velocities can reach 1 m/s and eddy currents extend at least 200 m into the water column. The eddy currents appear to cover the width of the sea, providing a pathway for rapid transport of marine organisms and other drifting material from one coast to the other. Interannual variability in the strength of the dipole is closely matched with variability in the strength of the wind jet. The dipole is observed to be quasi-stationary, although there is some evidence for slow eastward propagation—simulation of the dipole in an idealized high-resolution numerical model suggests that this is the result of self-advection. These and other recent in situ observations in the Red Sea show that the upper ocean currents are dominated by mesoscale eddies rather than by a slow overturning circulation.

Description: This work is supported by Award Nos. USA 00002, KSA 00011 and KSA 00011/02 made by King Abdullah University of Science and Technology (KAUST).

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1999

Description: Observations of diapycnal mixing rates are examined and related to diapycnal advection for both double-diffusive and turbulent regimes. The role of double-diffusive mixing at the site of the North Atlantic Tracer Release Experiment is considered. The strength of salt-finger mixing is analyzed in terms of the stability parameters for shear and double-diffusive convection, and a nondimensional ratio of the thermal and energy dissipation rates. While the model for turbulence describes most dissipation occurring in high shear, dissipation in low shear is better described by the salt-finger model, and a method for estimating the salt-finger enhancement of the diapycnal haline diffusivity over the thermal diffusivity is proposed. Best agreement between tracer-inferred mixing rates and microstructure based estimates is achieved when the salt-finger enhancement of haline flux is taken into account. The role of turbulence occurring above rough bathymetry in the abyssal Brazil Basin is also considered. The mixing levels along sloping bathymetry exceed the levels observed on ridge crests and canyon floors. Additionally, mixing levels modulate in phase with the spring-neap tidal cycle. A model of the dissipation rate is derived and used to specify the turbulent mixing rate and constrain the diapycnal advection in an inverse model for the steady circulation. The inverse model solution reveals the presence of a secondary circulation with zonal character. These results suggest that mixing in abyssal canyons plays an important role in the mass budget of Antarctic Bottom Water.

Description: This work was supported by contracts N00014-92-1323 and N00014-97-10087 of the Office of Naval Research and grant OCE94-15589 of the National Science Foundation.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution January 1988

Description: Analysis of vertical profiles of absolute horizontal velocity collected in January 1981, February 1982 and April 1982 in the central equatorial Pacific as part of the Pacific Equatorial Ocean Dynamics (PEQUOD) program, revealed two significant narrow band spectral peaks in the zonal velocity records, centered at vertical wavelengths of 560 and 350 stretched meters (sm). Both signals were present in all three cruises, but the 350 sm peak showed a more steady character in amplitude and a higher signal-to-noise ratio. In addition, its vertical scales corresponded to the scales of the conspicuous alternating flows generically called the equatorial deep jets in the past (the same terminology will be used here). Meridional velocity and vertical displacement spectra did not show any such energetic features. Energy in the 560 sm band roughly doubled between January 1981 and April 1982. Time lagged coherence results suggested upward phase propagation at time scales of about 4 years. East-west phase lines computed from zonally lagged coherences, tilted downward towards the west, implying westward phase propagation. Estimates of zonal wavelength (on the order of 10000 km) and period based on these coherence calculations, and the observed energy meridional structure at this vertical wavenumber band, seem consistent, within experimental errors, with the presence of a first meridional mode long Rossby wave packet, weakly modulated in the zonal direction. The equatorial deep jets, identified with the peak centered at 350 sm, are best defined as a finite narrow band process in vertical wavenumber (311-400 sm), accounting for only 20% of the total variance present in the broad band energetic background. At the jets wavenumber band, latitudinal energy scaling compared well with Kelvin wave theoretical values and a general tilt of phase lines downward towards the east yielded estimates of 10000-16000 km for the zonal wavelengths. Time-lagged coherence calculations revealed evidence for vertical shifting of the jets on interannual time scales. Interpretation of results in terms of single frequency linear wave processes led to inconsistencies, but finite bandwidth (in frequency and wavenumber) Kelvin wave processes of periods on the order of three to five years could account for the observations. Thus, the records do not preclude equatorial waves as a reasonable kinematic description of the jets.

Description: This research was supported by grant OCE-8600052 from the National Science Foundation, through the Woods Hole Oceanographic Institution.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2011

Description: Hurricane activity in the Northeastern Gulf of Mexico and its relationship to regional and large-scale climate variability during the Late Holocene is explored. A 4500-year record of hurricane-induced storm surges is developed from sediment cores collected from a coastal sinkhole near Apalachee Bay, Florida. Reconstructed hurricane frequency is shown to exhibit statistically significant variability with the greatest activity occurring between 2700 and 2400 years ago and the least activity between 1900 to 1600 years ago and after 600 years ago. Proxy records of stormrelevant climate variables contain similar timescales of variability and suggest both regional and large-scale mechanisms have influenced hurricane activity on centennial to millennial timescales. In particular, low-frequency migrations of the Loop Current may exercise control over regional hurricane activity by changing the thermal structure of the upper ocean and influencing the role of storm-induced upwelling on hurricane intensification. A new method for estimating the frequency of hurricanegenerated storm surges is presented and applied to Apalachee Bay, Florida. Multisite paleohurricane reconstructions from this region are developed, and the effects of geographic boundary conditions and temporal resolution on estimates of paleohurricane frequency are explored.

Description: Financial support provided by the American Meteorological Society, the National Science Foundation, the Bermuda Risk Prediction Initiative, the National Center for Airborne Laser Mapping, and the Coastal Ocean Institute.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1999

Description: An algorithm is developed for underwater acoustic signal processing with an array of hydrophones. With various acoustic signals coming from different directions, the maximum likelihood approach is used to estimate the source bearings and time series. Simulated annealing is used to implement the resulting time-domain beamformer. Broadband signals in spatially correlated noise are treated. Previous time-domain beamformers did not consider the correlation between random noise, and they did not use the concept of maximum likelihood, which is asymptotically optimal. We show that improved resolution can be achieved using this new method.

Description: Submitted in partial fulfillment of the requirements for the degree of Ocean Engineer at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 1988

Description: Two approaches to determining the ocean sound speed profile using measured acoustic modal eigenvalues are examined. Both methods use measured eigenvalues and mode dependent assumed values of the WKB phase integral as input data and use the WKB phase integral as a starting point for relating the index of refraction to depth. Inversion method one is restricted to monotonic or symmetric sound speed profiles and requires a measurement of the sound speed at one depth to convert the index of refraction profile to a sound speed profile. Inversion method two assumes that the sound speed at the surface and the minimum sound speed in the profile are known and is applicable to monotonic profiles and to general single duct sound speed profiles. For asymmetric profiles, inversion method two gives the depth difference between two points of equal sound speed in the portion of the profile having two turning points, and in the remainder of the profile it gives sound speed versus depth directly. A numerical implementation of the methods is demonstrated using idealized ocean sound speed profiles numerical experiments used to test the performance of the inversions using noisy data. The two methods are used to determine the sediment sound speed profiles in two shallow water waveguide models, and inversion method one is used to find the sediment sound speed profile using data from an experiment performed in the Gulf of Mexico.

Description: Funding through the Office of Naval Research Fellowship Program.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution January 1988

Description: This study focuses on the zonal weakening, eastern termination and seasonal variations of the Atlantic equatorial undercurrent (EUC). The main and most original contribution of the dissertation is a detailed analysis of the Atlantic EUC simulated by Philander and Pacanowski's (1986) general circulation model (GCM), which provides a novel description of the dynamical regimes governing various regions of a nonlinear stratified undercurrent. Only in a narrow deep western region of the simulation does one find an approximately inertial regime corresponding to zonal acceleration. Elsewhere frictional processes cannot be ignored. The bulk of the mid-basin model EUC terminates in the overlying westward surface flow while only a small fraction (the deeper more inertial layers) terminates at the eastern coast. In agreement with observations, a robust feature of the GCM not present in simpler models is the apparent migration of the EUC core from above the thermocline in the west to below it in the east. In the GCM, this happens because the eastward flow is eroded more efficiently by vertical friction above the base of the thermocline than by lateral friction at greater depths. This mechanism is a plausible one for the observed EUC. A scale analysis using a depth scale which decreases with distance eastwards predicts the model zonal transition between western inertial and eastern inertio-frictional regimes. Historical and recent observations and simple models of the equatorial and coastal eastern undercurrents are reviewed, and a new analysis of current measurements in the eastern equatorial Atlantic is presented. Although the measurements are inadequate for definitive conclusions, they suggest that Lukas' (1981) claim of a spring surge of the Pacific EUC to the eastern coast and a seasonal branching of the EUC into a coastal southeastward undercurrent may also hold for the Atlantic Ocean. To improve the agreement between observed and modelled strength of the eastern undercurrent, it is suggested that the eddy coefficient of horizontal mixing should be reduced in future GCM simulations.

Description: This work was supported by NSF grants OCE82-14771, OCE82-08744 and OCE85-14885.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2011

Description: Current underwater acoustic channel estimation techniques generally apply linear MMSE estimation. This approach is optimal in a mean square error sense under the assumption that the impulse response fluctuations are well characterized by Gaussian statistics, leading to a Rayleigh distributed envelope. However, the envelope statistics of the underwater acoustic communication channel are often better modeled by the K-distribution. In this thesis, by presenting and analyzing field data to support this claim, I demonstrate the need to investigate channel estimation algorithms that exploit K-distributed fading statistics. The impact that environmental conditions and system parameters have on the resulting distribution are analyzed. In doing so, the shape parameter of the K-distribution is found to be correlated with the source-to-receiver distance, bandwidth, and wave height. Next, simulations of the scattering behavior are carried out in order to gain insight into the physical mechanism that cause these statistics to arise. Finally, MAP and MMSE based algorithms are derived assuming K-distributed fading models. The implementation of these estimation algorithms on simulated data demonstrates an improvement in performance over linear MMSE estimation.

Description: This work was supported by the Office of Naval Research grant #N00014-05-10085 and the National Science Foundation grant #OCE-0519903.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution April 8, 1988

Description: Several experiments are presented in this thesis which examine methods to measure and monitor fluid flow from hydrothermal vent fields. Simultaneous velocity, temperature, and conductivity data were collected in the convective flow emanating from a hydrothermal vent field located at 10°S6'N, 103° 41'W on the East Pacific rise. The horizontal profiles obtained indicate that the flow field approaches an ideal plume in the temperature and velocity distribution. Such parameters as total heat flow and maximum plume height can be estimated using either the velocity or the temperature information. The results of these independent calculations are in close agreement, yielding a total heat flow from this vent site of 3.7 ± 0.8 MW and a maximum height of 150±10 m. The nonlinear effects of large temperature variations on heat capacity and volume changes slightly alter the calculations applied to obtain these values. In Guaymas Basin, a twelve day time series of temperature data was collected from a point three centimeters above a diffuse hydrothermal flow area. Using concurrent tidal gauge data from the town of Guaymas it is shown that the effects of tidar currents can be strong enough to dominate the time variability of a temperature signal at a fixed point in hydrothermal flow and are a plausible explanation for the variations seen in the Guaymas Basin temperature data. Theoretical examination of hot, turbulent, buoyant jets exiting from hydrothermal chimneys revealed acoustic source mechanisms capable of producing sound at levels higher than ambient ocean noise. Pressure levels and frequency generated by hydrothermal jets are dependent on chimney dimensions, fluid velocity and temperature and therefore can be used to monitor changes in these parameters over time. A laboratory study of low Mach number jet noise and amplification by flow inhomogeneities confirmed theoretical predictions for homogeneous jet noise power and frequency. The increase in power due to convected flow inhomogeneities, however, was lower in the near field than expected. Indirect evidence of hydrothermal sound fields (Reidesel et al., 1982; Bibee and Jacobson, 1986) showing anomalous high power and low frequency noise associated with vents is due to processes other than jet noise. On Axial Seamount, Juan de Fuca Ridge, high quality acoustic noise measurements were obtained by two hydrophones located 3 m and 40 m from an active hydrothermal vent, in an effort to determine the feasibility of monitoring hydrothermal vent activity through floW noise generation. Most of the noise field could be attributed to ambient ocean noise sources of microseisms, distant shipping and weather, punctuated by local ships and biological sources. Water/rock interface waves of local origin, were detected which showed high pressure amplitudes near the seafloor and, decaying with vertical distance, produced low pressures at 40 m above the bottom. Detection of vent signals was hampered by unexpected spatial non- stationarity due to shadowing effects of the caldera wall. No continuous vent signals were deemed significant based on a criterion of 90% probability of detection and 5% probability of false alarm. However, a small signal near 40 Hz, with a power level of 1x10-4 Pa2/Hz was noticed on two records taken near the Inferno black smoker. The frequency of this signal is consistent with predictions and the power level suggests the occurrence of jet noise amplification due to convected density inhomogeneities. Ambient noise from the TAG (Trans-Atlantic Geotraverse) hydrothermal area on the Mid-Atlantic Ridge near 26°N, in the frequency band 1-30 Hz at a range of 0.75-14 km from the site of an extremely active high temperature hydrothermal vent field (Rona, 1986) was examined. The ambient noise field exhibits great temporal and spatial variations attributed in part to typical ocean noise sources such as distant shipping and microseisms. Power spectral levels as measured at each of six ocean bottom hydrophones (OBH) were used to estimate the location of point sources of sound in the area, if any. The hydrothermal vent did not produce enough sound to be located as a point source using data from the OBH array. The only consistently identifiable point source found with the data set was generating sound in a 0.8-3.5 Hz bandwidth and located outside the median valley. It appears to be harmonic tremor associated with the tip of a ridge on the western side of the spreading axis and may be volcanic in origin.

Description: This work was supported by the WHOI/MIT Education Office, the Center for Analysis of Marine Systems, the National Science Foundation (grant OCE83-l0l75), NOAA National Sea Grant College Program Office, Dept. of Commerce under grant #NA86-AA-D-SG090, WHOI Sea Grant (R/6-l4), the Office of Naval Research grant #N0014-87-K-0007, and the NOAA Vents Program.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1988

Description: In previous Gulf Stream work (Hall and Bryden, 1985, Hall, 1985, 198GA, 198GB), a decomposition of multiple depth current records was developed which produced along- and cross-stream components. The cross-stream component was found to occasionally match lateral displacements of the Stream, as determined by temperature changes measured at the current meters. This study determined where within the meander pattern of the Gulf Stream the cross-stream velocity calculated from current meters at depth correctly predicted translations of the Gulf Stream as measured by satellite data. Additionally, the effects of recently quantified cross-stream velocities associated with the curvature of Gulf Stream meanders were analyzed.

Description: Funds for this work were provided by ONR contracts N00014-86-K-0751 and N00014-87-K-0007.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1988

Description: The highly energetic Agulhas Retroflection region south of the African continent lies at the junction of the South Indian, South Atlantic, and Circumpolar Oceans. A new survey of the Agulhas Retroflection taken in March 1985, plus historical hydrographic data, allow its dynamical and water-mass characteristics, and its role in exchanging mass, tracers, and vorticity between the three oceans, to be extensively characterized. The 1985 survey is composed of three independent, synoptic elements: a grid of closely-spaced, full-water-depth hydrographic stations (the first entirely full-water-column survey in this area), including several transects of the Agulhas and Agulhas Return Currents; a continuous survey of the path of the currents (the first such survey in the Agulhas); and a contemporaneous and relatively cloud-free sea surface temperature image derived from satellite infrared measurements. Mass transport balances within the closed grid boxes of the 1985 hydrographic survey provide information about current transport, recirculation (transport in excess of estimated returning interior ocean transport), and the overall Retroflection transport pattern. The current transport values exceed by as much as a factor of 1.5 the maximum interior transport computed from observed wind-stress curl and linear theory. Agulhas Current transports ranged from 56 to 95 x 106 m s-l at four 1985 transects crossing the current. Agulhas Return Current transports at the two 1985 transects were 54 and 65 x 106 m s-l. These transports are computed relative to 2400 dbar, which lies below the deep oxygen minimum emanating from the South Indian Ocean, and above the North Atlantic Deep Water salinity maximum. The current retroflected in two distinct branches in 1985, with a cold ring and a partially isolated warm recirculation cell found between the two branches. The satellite-derived sea surface temperature (SST) image, in agreement with the in situ measurements, showed that the cold ring lacked a cold SST anomaly; that the subsurface current path, as represented by a survey of the 15 C isotherm and 200 dbar surface intersection, was closely followed by a sharp front in sea surface temperature; and that most of the Agulhas's surface warm core retroflected upstream of the second retroflection branch. Anticyclonic curvature vorticity at sharp turns in the subsurface current path was found to exceed the maximum allowed by gradient wind balance, indicating that at these locations time-dependence and cross-frontal flow are important. The current's density field is found to meet necessary conditions for baroclinic and barotropic instability. These instability mechanisms may play a role in ring formation and current meandering. Top-to-bottom cross-stream spatial and isopycnal water-mass layering in the Agulhas Current, Agulhas Return Current, and associated rings are presented in two sets of sections, one contoured with pressure and the other with potential density as vertical coordinate. Temperature, salinity, oxygen, potential density and velocity sections are shown contoured versus pressure; and pressure, salinity, oxygen, and planetary potential vorticity are shown contoured versus potential density. These sections clearly illustrate water-mass structure both in space and relative to isopycnal surfaces. Strong salt, oxygen, and potential vorticity fronts on isopycnals in the upper -300 m across the Agulhas and Agulhas Return Current are observed, as are deep western boundary filaments of (i) salty, low oxygen water at intermediate depths traceable to Red Sea Water influences, and (ii) salty North Atlantic Deep Water close round the tip of Africa. The 1985 cold-core ring is the first cold-cored isolated feature to be observed within the Retroflection itself. Its transport was 64 x 106 m s-1, its integrated kinetic and available potential energy anomalies were 8.3 and 61 x 1015 J respectively, and its integrated planetary potential vorticity anomaly was 2.8 x 10-12 m-1 s-1. The potential vorticity flux associated with the exchange of 25 warm ring/cold ring pairs per year between the South Indian and Southern Oceans would balance the potential vorticity input by the wind to the entire South Indian Ocean. Interbasin flow of warm thermocline water (warmer than 8 C) from the South Indian to the South Atlantic Ocean is reconsidered in light of the 1985 hydrographic data. Thermocline water flow from the South Indian Ocean into the South Atlantic in the 1985 and historical observations is found to range from 2.8 to 〈9.6 x 106 m s-I. These values are less than the S;10 x 106 m s·1 needed to balance the Atlantic Ocean export of deep water, and implies that the deep water export is balanced in part by water colder than 8 C.

Description: Funding was provided by the Office of Naval Research under contract numbers NOOOI4-84-C-OI34 (NR083-400), NOOOI4-85-C-OOOl (NR083-004), and NOOOI4-87-K-0007 (NR083-004).

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2011

Description: Throughout this thesis we will discuss the theoretical background and empirical observation of a swell band shore normal flux divergence reversal. Specifically, we will demonstrate the existence and persistence of the energy flux divergence reversal in the nearshore region of Atchafalaya Bay, Gulf of Mexico, across storms during the March through April 2010 deployment. We will show that the swell band offshore component of energy flux is rather insignificant during the periods of interest, and as such we will neglect it during the ensuing analysis. The data presented will verify that the greatest flux divergence reversal is seen with winds from the East to Southeast, which is consistent with theories which suggest shoreward energy flux as well as estuarine sediment transport and resuspension prior to passage of a cold front. Employing the results of theoretical calculations and numerical modeling we will confirm that a plausible explanation for this phenomena can be found in situations where temporally varying wind input may locally balance or overpower bottom induced dissipation, which may also contravene the hypothesis that dissipation need increase shoreward due to nonlinear wave-wave interactions and maturation of the spectrum. Lastly, we will verify that the data presented is consistent with other measures collected during the same deployment in the Atchafalaya Bay during March - April 2010.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1998

Description: In practical applications with bathymetric sidescan sonars, the multipath reflections and other directional interferences are the key limiting factors for a better performance. This thesis proposes a new scheme to deal with the interferences using a multiple-row bathymetric sidescan sonar. Instead of smoothing the measurements over some time or angle intervals, which was previously widely investigated, we resolve the multipath interferences from the direct signal. Two approaches on signal direction-of-arrival DOA and amplitude estimation are developed, the correlated signal direction estimate CSDE for three-row systems and the ESPRIT-based method. These approaches are compared using different sonar data models, including a stochastic model from the statistical analysis on bottom scattering and a coherent model from the analysis on interference field; the simulations show the ESPRIT-based approach is quite robust at the angular separation of 100 between two sources and at the signal-to-noise ratio above 10dB except for highly coherent or temporally correlated signals, for which CSDE works very well. The computer simulation results and the discussions on practical algorithm implementation indicate the proposed scheme can be applied to a real multiple-row bathymetric sidescan sonar. With the capability to simultaneously resolve two or more directional signals, the new sonar model should work better for a wider variety of practical situations in shallow water with out significant increase of the system cost.

Description: Funding supporting my thesis research project was provided by the Office of Naval Research ONR.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution August 1999

Description: The thesis investigates the circulation at a 76-m deep study site on the southern flank of Georges Bank, a shallow submarine bank located between the deeper Gulf of Maine and the continental slope. Emphasis is placed on the vertical structure of the bottom boundary layer driven by the semi diurnal tides and the flow field's response to wind forcing. The observational analysis presented here is based on a combination of moored array and bottom tripod-mounted current, temperature, conductivity, and meteorological data taken between February and August 1995. Results from the bottom boundary layer analysis are compared to numerical model predictions for tidal flow over rough bottom topography. The flow response to wind forcing is examined and brought into context with the existing understanding of the wind-induced circulation in the Georges Bank region. Particular attention is given to the vertical distribution of the wind-driven currents, whose variation with background stratification is discussed and compared to observations from open ocean studies.

Description: The research presented in this thesis was generously supported by the National Science Foundation under grants OCE 93-13671 and OCE 96-32357 as part of the U.S. GLOBEC/Georges Bank Program.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1999

Description: A state-of-the-art, high-resolution ocean general circulation model is used to estimate the time-dependent global ocean heat transport and investigate its dynamics. The north-south heat transport is the prime manifestation of the ocean’s role in global climate, but understanding of its variability has been fragmentary owing to uncertainties in observational analyses, limitations in models, and the lack of a convincing mechanism. These issues are addressed in this thesis. Technical problems associated with the forcing and sampling of the model, and the impact of high-frequency motions are discussed. Numerical schemes are suggested to remove the inertial energy to prevent aliasing when the model fields are stored for later analysis. Globally, the cross-equatorial, seasonal heat transport fluctuations are close to +4.5 x 1015 watts, the same amplitude as the seasonal, cross-equatorial atmospheric energy transport. The variability is concentrated within 200 of the equator and dominated by the annual cycle. The majority of it is due to wind-induced current fluctuations in which the time-varying wind drives Ekman layer mass transports that are compensated by depth-independent return flows. The temperature difference between the mass transports gives rise to the time-dependent heat transport. The rectified eddy heat transport is calculated from the model. It is weak in the central gyres, and strong in the western boundary currents, the Antarctic Circumpolar Current, and the equatorial region. It is largely confined to the upper 1000 meters of the ocean. The rotational component of the eddy heat transport is strong in the oceanic jets, while the divergent component is strongest in the equatorial region and Antarctic Circumpolar Current. The method of estimating the eddy heat transport from an eddy diffusivity derived from mixing length arguments and altimetry data, and the climatological temperature field, is tested and shown not to reproduce the model’s directly evaluated eddy heat transport. Possible reasons for the discrepancy are explored.

Description: Funding for this research came from the Department of Defense under a National Defense Science and Engineering Graduate Fellowship. Financial support was also contributed by the National Science Foundation through grants #OCE-9617570 and #OCE-9730071, and the Tokyo Electric Power Company through the TEPCO/MIT Environmental Research Program. The author received partial support from an MIT Climate Modeling Fellowship, made possible by a gift from the American Automobile Manufacturers Association.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution August 1988

Description: In order to determine the efficacy of tomographic reconstructions of the ocean sound speed structure in improving acoustic field predictions for source localization, a 150 km by 350 km volume of ocean 3000 meters deep was synthetically modeled to be similar to the Gulf Stream system, including an eddy and a front. The features were Gaussian, with the eddy's maximum sound speed perturbation being 10ms-1 and the front's maximum perturbation 15ms-1. Two vertical slices through this system were inverted in a synthetic tomography experiment using linear optimal estimation theory. Inversions were also performed using XSV and satellite sea surface temperature data. Gaussian fits to the reconstructed features were constructed for use with a three dimensional raytrace program (HARPO). Three dimensional rays were propagated both through the reconstructions and the original model. Travel time versus intensity (transmission loss) for the eigenrays was used as a basis for intercomparison. Tomographic results showed good reconstruction for a first iteration of the inversion, but inadequate vertical resolution. Iterations and the use of more refractive eigenrays are needed for improvement of the reconstruction, especially for the front. Reconstructed results for the acoustic field should improve conventional beamforming, but are probably inadequate for matched field processing.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2009

Description: Interaction between the Antarctic Circumpolar Current and the continental slope/shelf in the Marguerite Bay and west Antarctic Peninsula is examined as interaction between a wind-driven channel flow and a zonally uniform slope with a bay-shaped shelf to the south. Two control mechanisms, eddy advection and propagation of topographic waves, are identified in barotropic vortex-escarpment interactions. The two mechanisms advect the potential vorticity (PV) perturbations in opposite directions in anticyclone-induced interactions but in the same direction in cyclone-induced interactions, resulting in dramatic differences in the two kinds of interactions. The topographic waves become more nonlinear near the western(eastern if in the Northern Hemisphere) boundary of the bay, where strong cross-escarpment motion occurs. In the interaction between a surface anticyclone and a slope penetrating into the upper layer in a two-layer isopycnal model, the eddy advection decays on length scales on the order of the internal deformation radius, so shoreward over a slope that is wider than the deformation radius, the wave mechanism becomes noticeably significant. It acts to spread the cross-isobath transport in a much wider range while the transport directly driven by the anticyclone is concentrated in space. A two-layer wind-driven channel flow is constructed to the north of the slope in the Southern Hemisphere, spontaneously generating eddies through baroclinic instability. A PV front forms in the first layer shoreward of the base of the topography due to the lower-layer eddy-slope interactions. Perturbed by the jet in the center of the channel, the front interacts with the slope/shelf persistently yet episodically, driving a clockwise mean circulation within the bay as well as crossisobath transport. Both the transports across the slope edge and out of the bay are comparable with the maximum Ekman transport in the channel, indicative of the significance of the examined mechanism. The wave-boundary interaction identified in the barotropic model is found essential for the out-of-bay transport and responsible for the heterogeneity of the transport within the bay. Much more water is transported out of the bay from the west than from the east, and the southeastern area is the most isolated region. These results suggest that strong out-of-bay transport may be found near the western boundary of the Marguerite Bay while the southeastern region is a retention area where high population of Antarctic krill may be found.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution August 1984

Description: This thesis addresses several aspects of the problem of determining the effect of the low-frequency eddy variability on the mean circulation of the Western North Atlantic. A framework for this study is first established by scale analysis of the eddy and mean terms in the mean momentum, vorticity, and heat balances in three regions of the Western North Atlantic -- the northern recirculation, the southern recirculation, and the mid-ocean. The data from the last decade of field experiments suggest somewhat different conclusions from the earlier analysis of Harrison (1980). In the momentum balance we confirm that the eddy terms are negligible compared to the lowest order mean geostrophic balance. The eddy term may be an 0(1) term in the vorticity balance only in the northern recirculation region where the mean flow is anisotropic. In the mean heat balance, if the mean temperature advection is scaled using the thermal wind relation, then the eddy heat flux is negligible in the mid-ocean, but it may be important in the recirculation areas. For all the balances the eddy terms are comparable to or an order of magnitude larger than the mean advective terms. We conclude from the scale analysis that the eddy field is most likely to be important in the Gulf Stream recirculation region. These balances are subsequently examined in more detail using data from the Local Dynamics Experiment (LDE). Several inconsistencies are first shown in McWilliams' (1983) model for the mean dynamical balances in the LDE. The sampling uncertainties do not allow us to draw conclusions about the long-term dynamical balances. However, it is shown that if we assume that the linear vorticity balance holds between the surface and the thermocline for a finite record, then the vertical velocity induced by the eddy heat flux divergence is non-zero. The local effect of the mesoscale eddy field on the mean potential vorticity distribution of the Gulf Stream recirculation region is determined from the quasigeostrophic eddy potential vorticity flux. This flux is calculated by finite difference of current and temperature time series from the Local Dynamics Experiment. This long-term array of moorings is the only experimental data from which the complete eddy flux can be calculated. The total eddy flux is dominated by the term due to the time variation in the thickness of isopycnal layers. This thickness flux is an order of magnitude larger than the relative vorticity flux. The total flux is statistically significant and directed 217° T to the southwest with a magnitude of 1.57 x 10 -5 cm/2s. The direction of the eddy flux with respect to the mean large scale potential vorticity gradient from hydrographic data indicates that eddies in this region tend to reduce the mean potential vorticity gradient. The results are qualitatively consistent with numerical model results and with other data from the Gulf Stream recirculation region. We find that the strength of the eddy transfer in the enstrophy cascade is comparable to the source terms in the mean enstrophy balance. The Austauch coefficient for potential vorticity mixing is estimated to be 0(107cm2/sec). An order of magnitude estimate of the enstrophy dissipation due only to the internal wave field shows that other processes must be important in enstrophy dissipation. The measured eddy potential vorticity fluxes are compared to the linear stability model of Gill, Green, and Simmons (1974). An earlier study (Hogg, 1984) has shown agreement between the empirical orthogonal modes of the data and the predicted wavenumbers, growth rates, and phase speeds of the most unstable waves. However, we show substantial disagreement in a comparison of the higher moments the eddy heat and potential vorticity fluxes. Because the critical layer of the model is located near the surface, the model predicts that most of the eddy potential vorticity and eddy heat flux should occur within about 300 meters of the surface. The data show much greater deep eddy heat flux than predicted by the model. It is suggested that the unstable modes in the ocean have a longer vertical scale because of the reduction in the buoyancy frequency near the surface. The evidence for in situ instability is also examined in the decay region of the Gulf Stream from an array of current and temperature recorders. Although there is vertical phase propagation in the empirical orthogonal modes for some of the variables at some of the moorings, there is not much evidence for a strong ongoing process of wave generation.

Description: This research has been conducted under NSF contract numbers OCE 77-19403, ATM 79-21431, and OCE 82-00154.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution May 1985

Description: The acoustic properties of marine sediments have a direct effect on the propagation of sound in the ocean. In the frequency range of interest (50 - 500 Hz) the sediment can be modelled as a fluid. Assuming horizontal stratification of the ocean bottom, the acoustic parameters of interest are the compressional wave speed, the compressional wave attenuation and density as a function of depth. An inverse method based on a perturbation technique is presented in this thesis for the determination of these parameters. A monochromatic source experiment is proposed because of the desirability of such an experiment for determining the acoustic properties of an anelastic medium. The input information is the plane wave reflection coerricent as a function of the angle of incidence at a fixed frequency. A nonlinear integral equation relating the variations of these acoustic properties from a known reference value to the plane wave reflection coefficient is derived. This is then linearised using the Born approximation. The region of validity of the Born approximation is derived and based on this the optimum angular aperture for the input data is obtained. The linearised integral equation is a Fredholm integral equation of the first kind. An acceptable stable solution of the integral equation is obtained by imposing a priori constraints on the solution. The inversion method is tested using synthetic data and inversions are carried out for various examples of the attenuation coefficient profile and the sound speed profile. The results obtained with noise free data show good agreement between the true profiles and the reconstructed profiles. The resolution obtainable with the data set is studied using the resolving power theory of Backus and Gilbert and the inversion method is shown to provide adequate resolution. The effect of additive noise in data is examined and inversions performed with noisy data yielded stable acceptable results.

Description: I acknowledge the financial support provided by the education office in the Woods Hole Oceanographic Institution and the Office of Naval Research.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution May 1986

Description: The dual issues of modal decomposition for tonal sound fields and the temporal coherence of the modal amplitudes are investigated for the case of the central Arctic sound channel at very low frequencies (15-80 Hz). A detailed study of the Arctic modal structure for these frequencies reveals the central role played by the strong Arctic surface duct. The performance of each of four different modal beamforming algorithms when applied to the vertical array deployed during the FRAM IV Arctic Acoustic Experiment is analyzed. A multiple beam (or decoupled beam) least squares processor produces the most acceptable results for Arctic conditions. The modal decomposition is sensitive to vertical array tilt caused by hydrodynamic drag; a technique for its estimation from the acoustie data is developed. Tonal data taken from both the horizontal and vertical arrays deployed during FRAM IV is analyzed. Horizontal array results confirm the modal amplitudes generated from vertical array data. The rough surface scattering from the ice canopy places an upper limit of 40 Hz on efficient surface duct propagation. Attenuation measurements for the first mode show excellent agreement with predictions made for ice scattering using the method of small perturbations and experimental ice statistics. The high levels of coherence observed (O.95 to 0.99) show that tonal signal propagation in the Arctic channel is essentially deterministic for time periods well in excess of one hour. The various modes may then be considered to maintain a constant phase relationship over time.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution December 1999

Description: A new, global inversion is used to estimate the large scale oceanic circulation based on the World Ocean Circulation Experiment and Java Australia Dynamic Experiment hydrographic data. A linear inverse "box" model is used to combine consistently the transoceanic sections. The circulation is geostrophic with an Ekman layer at the surface and oceanic layers defined by neutral surfaces. Near-conservation of mass, salt and top-to-bottom silica is required and, in addition, heat and the phosphate-oxygen combination (170[P04]+[02]) are conserved in layers that are not in contact with the surface. A globally-consistent solution is obtained for a depth-independent adjustment to the thermal wind field, freshwater flux divergenees, the Ekman transport, and the advective and diffusive dianeutral fluxes between layers. A detailed error budget permits calculation of statistical uncertainties, taking into account both the non-resolved part of the solution and the systematic errors due to the temporal oceanic variability. The estimated water mass transports during the WOCE period (1985-1996) are generally similar to previous published estimates. However, important differences are found. In particular, the inflow of bottom waters into the Pacific Ocean is smaller than in most previous estimates. Utilization of property anomaly conservation constraints allows the estimation of significant dianeutral diffusivities in deep layers, with a global average of 3 ± lcm2s- 1 north of 30°S. Dianeutral transfers indicate that about 20 Sv of bottom water is formed in the Southern Ocean. Significant ocean-atmosphere heat fluxes are found, with a global heating of 2.3 ± 0.4PW in the tropical band and a corresponding cooling at high latitudes. The signature of a large-scale average export production is found for nutrients in several temperate regions. Despite the large uncertainties, the production magnitudes are consistent with independent measurements from sediment traps and isotopic data. Net nutrient sources or sinks are found in several regions, suggesting either transport of dissolved organic matter or a seasonal alias. Oxygen indicates large exchanges with the atmosphere, with intake at high latitudes and outgassing/remineralization at low latitudes.

Description: This work was supported in part by the Jet Propulsion Laboratory/CALTECH (contract #958125), and by gifts from Ford, General Motors, and Daimler-Chrysler to MIT's Climate Modelling Initiative.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution December 1991

Description: Breaking waves dissipate energy, transfer momentum from the wind to surface currents and breaking enhances the transfer of gas and mass across the air-sea interface. Breaking waves are believed to be the dominant source of sea surface sound at frequencies greater than 500 Hz and the presence of breaking waves on the ocean surface has been shown to enhance the scattering of microwave radiation. Previous studies have shown that breaking waves can be detected by measuring the microwave backscatter and acoustic radiation from breaking waves. However, these techniques have not yet proven effective for studying the dynamics of breaking. The primary motivation for the research presented in this thesis was to determine whether measurements of the sound generated by breaking waves could be used to quantitatively study the dynamics of the breaking process. Laboratory measurements of the microwave backscatter and acoustic radiation from two-dimensional breaking waves are described in Chapter 2. The major findings of this Chapter are: 1) the mean square acoustic pressure and backscattered microwave power correlate with the wave slope and dissipation for waves of moderate slope, 2) the mean square acoustic pressure and backscattered microwave power correlate strongly with each other, and 3) the amount of acoustic energy radiated by an individual breaking event scaled with the amount of mechanical energy dissipated by breaking. The observed correlations with the mean square acoustic pressure are only relevant for frequencies greater than 2200 Hz because lower frequencies were below the first acoustic cut-off frequency of the wave channel. In order to study the lower frequency sound generated by breaking waves another series of two-dimensional breaking experiments was conducted. Sound at frequencies as low as 20 Hz was observed and the mean square acoustic pressure in the frequency band from 20 Hz-l kHz correlated strongly with the wave slope and dissipation. A characteristic low frequency signal was observed immediately following the impact of the plunging wave crest. The origin of this low frequency signal was found to be the pulsating cylinders of air which are entrained by the plunging waves. The pulsation frequency correlated with both the wave slope and dissipation. Following the characteristic constant frequency signal, approximately 0.25 s after the initial impact of the plunging crest, another low frequency signal was typically observed. These signals were generally lower in frequency initially and then increased in frequency as time progressed. To determine if three-dimensional effects were important in the sound generation process and to measure the sound beneath larger breaking waves a series of experiments was conducted in a large multi-paddle wave basin. Three-dimensional breaking waves were generated and the sound produced by breaking was measured in the frequency range from 10 Hz to 20 kHz. The observed sound spectra showed significant increases in level across the entire bandwidth from 10 Hz to 20 kHz and the spectra sloped at -5 to-6 dB per octave at frequencies greater than 1 kHz. The mean square acoustic pressure in the frequency band from 10 Hz to 150 Hz correlated with the wave amplitude similar to the results obtained in the two-dimensional breaking experiments. Large amplitude low frequency spectral peaks were observed approximately 0.75 s after the initial impact of the plunging crests. It was postulated that the low frequency signals observed some time after the initial impact of the plunging crests for both the two and three-dimensional breakers were caused by the collective oscillation of bubble clouds. Void fraction measurements taken by Eric Lamarre were available for five breaking events and therefore the average sound speed inside the bubble clouds and their radii were known. Using this information the resonant frequencies of a two-dimensional cylindrical bubble cloud of equal radius and sound speed were calculated. The frequencies of the observed signals matched closely with the calculated resonant frequencies of the first and second mode of the two-dimensional cylindrical bubble cloud. The close agreement supports the hypothesis that the low frequency signals were produced by the collective oscillation of bubble clouds. In Chapter 4 a model of the sound produced by breaking waves is presented which uses the sound radiated by a single bubble oscillating at its linear resonant frequency and the bubble size distribution to estimate the sound spectrum. The model generates a damped sinusiodal pulse for every bubble formed, as calculated from the bubble size distribution. If the range to the receiver is known then the only unknown parameters are ε, the initial fractional amplitude of the bubble oscillation and L, the dipole moment arm or twice the depth of the bubble below the free surface. It was found that if the product εxL is independent of the bubble radius the model reproduces the shape and magnitude of the observed sound spectrum accurately. The success of the model implies that it may be possible to calculate the bubble size distribution from the sound spectrum. The model was validated using data from experiments where the breaking events were small scale gently spilling waves (Medwin and Daniel, 1990).

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1985

Description: From October, 1982 to October, 1983 a current meter mooring reaching from the bottom into the thermocline was deployed for the first time in the Gulf Stream at 68°W. The temperatures, pressures, and velocities at the uppermost instrument indicate the Gulf Stream moved back and forth across the mooring site, so that the entire Stream was sampled in time; hence the data may be used to examine horizontal as well as vertical structure of the Stream. The two key points to the success of the analysis are: 1)the well-defined relationship between temperature and cross-stream distance in the thermocline, enabling the use of the former as a horizontal coordinate; and 2)a daily-changing definition of Gulf Stream flow direction based on the shear between the thermocline and 2000 m depth. Time-series of daily-rotated velocities may be used to calculate empirical orthogonal functions for the long- and cross-stream vertical structures, which are decoupled and are respectively baroclinic and barotropic. Using the inferred horizontal coordinate one can estimate mass, momentum and kinetic energy fluxes for four individual events when the entire Stream swept by the mooring. The results agree well with historical data. Bryden's (1980) method has been used to calculate vertical velocities from the temperature equation; the resulting time-series of w are visually coherent throughout the water column and their vertical amplitude structure is reminiscent of that for a two-layer system. The rms vertical velocities are large (0(.05 cm/s)), and these as well as other estimates have been used to explore the validity of the quasi-geostrophic approximation at the mooring site. The Rossby number for the thermocline flow is about 0.3, and for the deep flow is ≤ 0.1. The entire data set may also be used to construct a horizontal and vertical profile of velocity in the Gulf Stream, from which a cross-section of the mean potential vorticity can be produced. The latter shares many common feature with cross-sections from past work for a nearby site, as well as analogous data from a three-layer numerical model, thus suggesting that they are robust features of Gulf Stream-like currents. These features are, in particular, a strong jump from low to high values crossing the Stream from south to north; and a change in the sign of the potential vorticity gradient on isothermal surfaces for T 〉 12°C. To complement the analysis of the observational data, a set of diagnostic calculations has been performed on an eddy-resolving qeneral circulation model, to provide a complete picture of the kinetic energy budgets of the free jet and its environs. It is found that the downstream convergence of kinetic energy in the decelerating jet is balanced primarily by an ageostrophic flow against the pressure gradient, which in turn implies some conversion of kinetic to available potential energy in the region. Energetic analysis of the observations as well as the numerical data suggests barotropic and baroclinic instabilities may be equally important to the kinetic energy budgets in the Stream. Because there is but one mooring, the dynamics governing the fluctuations remain elusive. Nonetheless, a kinematic framework is proposed, which is consistent with the data and accounts for a variety of unusual features that arise in the original analysis (for example, distinct asymmetries in the four Gulf Stream crossings, and the rather large vertical velocities). It is sugqested that the data we are now capable of collecting is proffering fundamentally new attributes of the Gulf Stream, which must be included and accounted for in future theoretical work.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution December 1980

Description: Data from the COBOLT experiment, which investigated the first 12 km off Long Island's south shore, are analyzed and discussed. Moored current meter records indicate that the nearshore flow field is strongly polarized in the alongshore direction and its fluctuations are well correlated with local meteorological forcing. Complex empirical orthogonal function analysis suggests that subtidal velocity fluctuations are barotropic in nature and are strongly influenced by bottom friction. Wind-related inertial currents were observed within the coastal boundary layer (CBL) under favorable meteorological and hydrographical conditions. The magnitude of these oscillations increases with distance from shore, and they display a very clear 180° phase difference between surface and bottom layers. Nearshore inertial oscillations of both velocity and salinity records appear to lead those further seaward, suggesting local generation and subsequent radiation away from the coast. The response of the coastal zone to impulsive wind forcing is discussed using simple slab and two-layer models, and the behavior of the nearshore current field examined. The major features of the observed inertial motions are in good qualitative agreement with model predictions. It is found that, in a homogeneous domain, the coastal boundary condition effectively prohibits inertial currents over the entire coastal zone. In the presence of stratification the offshore extent of this prohibition is greatly reduced and significant inertial currents may occur within one or two internal deformation radii of the coast. The "coastal effect", in the form of surface and interfacial waves which propagate away from the coast, modifies the "pure" inertial response as it would exist far from shore. The kinematics of this process is such that a 180° phase difference between currents in the two layers is characteristic of the entire coastal zone even before the internal wave has had time to traverse the CBL. It is also suggested that, for positions seaward of several internal deformation radii, interference between the surface and internal components of the coastal response will cause maximum inertial amplitudes to occur for t 〉 x/c2, where c2 is the phase speed of the internal disturbance. The hydrographic structure of the CBL is observed to undergo frequent homogenization. These events are related to both advective and mixing processes. Horizontal and vertical exchange coefficients are estimated from the data, and subsequently used in a diffusive model which accurately reproduces the observed mean density distribution in the nearshore zone. Dynamic balance calculations are performed which indicate that the subtidal cross-shore momentum balance is very nearly geostrophic. The calculations also suggest that the longshore balance may be reasonably represented by a steady, linear equation of motion which includes surface and bottom stresses. Evidence is presented which shows that variations in the longshore wind-stress component are primarily responsible for the energetic fluctuations in the sea surface slope along Long Island. Depth-averaged velocities characteristically show net offshore transport in the study area, and often display dramatic longshore current reversals with distance from shore. These observations are interpreted in terms of a steady circulation model which includes realistic nearshore topography. Model results suggest that longshore current reversals within the CBL may be limited to the eastern end of Long Island, and that this unusual flow pattern is a consequence of flow convergence related to the presence of Long Island Sound.

Description: This work was supported by the Department of Energy through contract no. DE-AC02-EVI0005 entitled Coastal-Shelf Transport and Diffusion.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2001

Description: Matched-field methods concern estimation of source location and/or ocean environmental parameters by exploiting full wave modeling of acoustic waveguide propagation. Typical estimation performance demonstrates two fundamental limitations. First, sidelobe ambiguities dominate the estimation at low signal-to-noise ratio (SNR), leading to a threshold performance behavior. Second, most matched-field algorithms show a strong sensitivity to environmental/system mismatch, introducing some biased estimates at high SNR. In this thesis, a quantitative approach for ambiguity analysis is developed so that different mainlobe and sidelobe error contributions can be compared at different SNR levels. Two large-error performance bounds, the Weiss-Weinstein bound (WWB) and Ziv-Zakai bound (ZZB), are derived for the attainable accuracy of matched-field methods. To include mismatch effects, a modified version of the ZZB is proposed. Performance analyses are implemented for source localization under a typical shallow water environment chosen from the Shallow Water Evaluation Cell Experiments (SWellEX). The performance predictions describe the simulations of the maximum likelihood estimator (MLE) well, including the mean square error in all SNR regions as well as the bias at high SNR. The threshold SNR and bias predictions are also verified by the SWellEX experimental data processing. These developments provide tools to better understand some fundamental behaviors in matched-field performance and provide benchmarks to which various ad hoc algorithms can be compared.

Description: Financial support for my research was provided by the Office of Naval Research and the WHOI Education Office.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2010

Description: In a stratified rotating fluid, frictionally driven circulations couple with the buoyancy field over sloping topography. Analytical and numerical methods are used to quantify the impact of this coupling on the vertical circulation, spindown of geostrophic flows, and the formation of a shelfbreak jet. Over a stratified slope, linear spindown of a geostrophic along-isobath flow induces cross-isobath Ekman flows. Ekman advection of buoyancy weakens the vertical circulation and slows spindown. Upslope (downslope) Ekman flows tend to inject (remove) potential vorticity into (from) the ocean. Momentum advection and nonlinear buoyancy advection are examined in setting asymmetries in the vertical circulation and the vertical relative vorticity field. During nonlinear homogeneous spindown over a flat bottom, momentum advection weakens Ekman pumping and strengthens Ekman suction, while cyclonic vorticity decays faster than anticyclonic vorticity. During nonlinear stratified spindown over a slope, nonlinear advection of buoyancy enhances the asymmetry in Ekman pumping and suction, whereas anticyclonic vorticity can decay faster than cyclonic vorticity outside of the boundary layers. During the adjustment of a spatially uniform geostrophic current over a shelfbreak, coupling between the Ekman flow and the buoyancy field generates Ekman pumping near the shelfbreak, which leads to the formation of a jet. Scalings are presented for the upwelling strength, the length scale over which it occurs, and the timescale for jet formation. The results are applied to the Middle Atlantic Bight shelfbreak.

Description: Funding for my research and education was provided by MIT EAPS, the WHOI Academic Programs O ce and the MIT Presidential Fellowship. Financial assistance from the Houghton Fund is also acknowledged.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2010

Description: The Labrador Sea, as one of a few places of deep water formation, plays an important role in the Meridional Overturning Circulation. While the interior of the Labrador Sea, where the deepest convection takes place, is known to experience variability on time scales ranging from days to decades, little is known about the variability of the other components of the Labrador Sea circulation - the boundary current system and the eddies that connect it with the interior. Using various types of in situ data combined with the surface flux and satellite altimetry data products, I studied the variability of both the boundary current system and the eddies on different time scales as well as their influence on the post-convective restratification of the Labrador Sea interior. The analysis presented in the thesis supports the result of the previous theoretical studies that argue that lateral fluxes, driven by the boundary current/interior gradients, play an important role in the post-convective restratification of the Labrador Sea. I found that both components of the boundary current, the surface West Greenland Current and the subsurface Irminger Current, have a strong seasonal cycle. In the spring both the West Greenland and Irminger Currents are colder and fresher than in the fall. However, the West Greenland Current is faster and thicker in the spring while the Irminger Current is the fastest and thickest in the fall. My analysis suggests that the observed seasonal changes in the velocity are primarily due to the baroclinic component of the current while the barotropic component remains nearly unchanged. The Subpolar Gyre, and the Labrador Sea in particular, have experienced a decline in the circulation accompanied by the warming of the water column over the last decades. I found that a similar trend is seen in the West Greenland Current system which slowed down from 1992 to 2004, primarily due to a decrease in the barotropic flow. At the same time, the subsurface Irminger Current has become warmer, saltier, and lighter, something that is also reflected in the properties of the eddies. Two years exhibited pronounced anomalies: in 1997 and 2003 the velocity, temperature and salinity of the Irminger Current abruptly increase with respect to the overall trend. Finally, I discuss the impacts of the boundary current changes on the lateral fluxes that are responsible for the restratification of the Labrador Sea and the properties of the interior.

Description: The financial support for my research came from the Academic Programs Office and from the NSF grants OCE-0424492 and OCE-0137023.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2010

Description: The variability in the DWBC, its connection to the forcing in the northern North Atlantic and interaction with the Gulf Stream were explored from a combination of remote sensing and in-situ measurements in the western North Atlantic. Using satellite altimetry and Sea Surface Temperature (SST) we found evidence of the relation between changes in the Gulf Stream path and the variability in the temperature and velocity fields in the Slope Water. This relation was such that southward shifts of the main axis of the Gulf Stream were preceded by cold temperature anomalies and intensification of the southwestward flow. The analysis of 5.5 years of moored CTD and horizontal velocity data in the DWBC at 69°W recorded during the period 2002-2008, showed that the variability along the DWBC is linked to changes in the dense water formation regions. The evolution of potential vorticity (PV) at the mooring site, characterized by a transition from deep to upper Labrador Sea Water (LSW), was similar to that observed in the Labrador Sea 6 to 9 years earlier, and imply spreading rates for the LSW that varied over time from 1.5 to 2.5cm/s. The time dependence of the spreading rates was in good agreement with changes in the strength of the DWBC at the mooring site. The evolution of the DWBC transport was explored in more detail from a 5-element moored array, also at 69°W. The results, for the period of 2004-2008, were consistent with the single mooring analysis. The variability measured from the array showed that upper, intermediate and deep water mass layers expand and contract at each other’s expense, leading to alternating positive and negative PV anomalies at the upper-LSW, deep-LSW and Overflow Water (OW). Larger DWBC transports were associated with enhanced presence of recently ventilated upper-LSW and OW, rather than deep-LSW. The relative contribution of the different water masses to the observed circulation was investigated by inverting individual PV anomalies isolated from the observations. We found that changes in the depth-integrated circulation were mostly driven by changes in the OW.

Description: During the first two years, my time here was funded through a La Caixa Foundation fellowship as well as an MIT presidential fellowship. Financial support for my thesis work was provided by National Science Foundation grants OCE-0241354 and OCE-0726720. Support for traveling to meetings was also provided by the MIT Houghton Fund and the WHOI Academic Programs Office.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2000

Description: A channel characterization experiment for the underwater acoustic communication channel was carried out at Scripps Pier in May 1999. The experiment investigated acoustic transmission in very shallow water and breaking waves. In analyzing the data, several questions arose. The majority of the acoustic channel probe data was corrupted by crosstalk in the receiver array cable. This thesis investigates methods to correct for the effects of the crosstalk, to attempt to recover the channel probe data. In selected regions, the crosstalk could be removed quite effectively using a linear least-squares method to estimate the crosstalk coefficients. The bulk of the data could not be corrected, however, primarily due to crosstalk from a receiver channel which was not recorded, and hence could not be well estimated. A second question addressed by this thesis is concerned with acoustic propagation in shallow water under bubble clouds. The breaking waves injected air deep into the water column. The resulting bubble clouds heavily attenuated acoustic signals, effectively causing total dropouts of the acoustic communication channel. Due to buoyancy, the bubbles gradually rise, and the communication channel clears. The channel clearing was significantly slower than predicted by geometric ray acoustic propagation models, however. Proposed explanations included secondary, unobserved, breaking events causing additional bubble injection; delayed rising of bubbles due to turbulent currents; or failure of the geometric ray model due to suppression by bubble clouds of acoustic signals which are not along the geometric ray paths. This thesis investigated the final hypothesis, modeling the acoustic propagation in Scripps Pier environment, using the full wave equation modeling package OASES. It was determined that the attenuation of the propagating acoustic signal is not accurately predicted by the bubble-induced attenuation along the geometric ray path.

Description: For financial support, thanks to the National Science Foundation for funding me on a Graduate Research Fellowship, and thanks to the WHOI Education Office for supplementing that fellowship.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution January 1987

Description: This thesis presents the theoretical and computational underpinnings of a novel approach to the determination of the acoustic parameters of the ocean bottom using a monochromatic source. The problem is shown to be equivalent to that of the reconstruction of the potential in a Schrodinger equation from the knowledge of the plane-wave reflection coefficient as a function of vertical wavenumber, r(kz) for all real positive k z. First, the reflection coefficient is shown to decay asymptotically at least as fast as (1/kz2) for large kz and is therefore inteqrable. The Gelfand-Levitan inversion procedure is extended to include the case of basement velocity higher than the velocity of sound in water. The neglect of bound states is shown to be justified in both clayey silt and silty clay at the 220 Hz frequency of operation. Three methods for the numerical solution of the integral equation are investigated. The first one is an "Improved Born approximation" wherein the solution is given as a series expansion the first term of which is the Born approximation while the second term represents a substantial and yet easy to implement improvement over Born. The two other methods are based on a discretization of the Gelfand-Levitan integral equation, and both avoid a matrix inversion: one by employing a recursive procedure, and the other by coupling the Gelfand-Levitan equation with a partial differential equation. Bounds are obtained on errors in the solution due either to discretization or to data inaccuracy. These methods are tested on synthetic data obtained from known geoacoustic models of the ocean bottom. Results are found to be very accurate particularly at the top of the sediment layer with resolution of less than the wavelength of the acoustic source in the water. Several effects are investigated, such as sampling, attenuation, and noise. Also examined is the gradual restriction of the reflection coefficient to a finite range of vertical wave numbers and the consequent progressive deterioration of the reconstruction. The analysis shows how to reconstruct velocity profiles in the presence of density variation when the experiment is conducted at two frequencies. Our results provide a good understanding of the issues involved in conducting a monochromatic deep ocean bottom experiment and constitute a promising technique for processing the experimental data when it becomes available.

Description: Support for this thesis was provided in part by the education office at WHOI and by the Office of Naval Research.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2002

Description: One of the major problems in wireless communications is compensating for the time-varying intersymbol interference (ISI) due to multipath. Underwater acoustic communications is one such type of wireless communications in which the channel is highly dynamic and the amount of ISI due to multipath is relatively large. In the underwater acoustic channel, associated with each of the deterministic propagation paths are macro-multipath fluctuations which depend on large scale environmental features and geometry, and micro-multipath fluctuations which are dependent on small scale environmental inhomogeneities. For arrivals which are unsaturated or partially saturated, the fluctuations in ISI are dominated by the macro-multipath fluctuations resulting in correlated fluctuations between different taps of the sampled channel impulse response. Traditional recursive least squares (RLS) algorithms used for adapting channel equalizers do not exploit this structure. A channel subspace post-filtering algorithm that treats the least squares channel estimate as a noisy time series and exploits the channel correlation structure to reduce the channel estimation error is presented. The improvement in performance of the algorithm with respect to traditional least squares algorithms is predicted theoretically, and demonstrated using both simulation and experimental data. An adaptive equalizer structure that explicitly uses this improved estimate of the channel impulse response is discussed. The improvement in performance of such an equalizer due to the use of the post-filtered estimate is also predicted theoretically, and demonstrated using both simulation and experimental data.

Description: This research was supported by an ONR Graduate Traineeship Award Grant #N00014-00-10049.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2009

Description: A confluence of several coastal oceanographic features creates an acoustically interesting region with high variability along the New England Shelfbreak. Determining the effect of the variability on acoustic propagation is critical for sonar systems. In the Nantucket Shoals area of the Middle Atlantic Bight, two experiments, the New England Shelfbreak Tests (NEST), were conducted in May and June, 2007 and 2008, to study this variability. A comprehensive climatology of the region along with the experimental data provided detailed information about the variability of the water column, particularly the temperature and sound speed fields. Empirical orthogonal function (EOF) analysis of the ocean sound speed field defined a set of perturbations to the background sound speed field for each of the NEST Scanfish surveys. Attenuation due to bottom sediments is the major contributor of transmission loss in the ocean. In shallow water, available propagation paths most often include bottom interaction. Perturbations in the ocean sound speed field can cause changes in the angle of incidence of sound rays with the bottom, which can result in changes to the amount of sound energy lost to the bottom. In lieu of complex transmission loss models, the loss/bounce model provides a simpler way to predict transmission loss changes due to perturbations in the background sound speed field in the ocean. Using an acoustic wavenumber perturbation method, sound speed perturbations, defined by the ocean EOF modes, are translated into a change in the horizontal wavenumber, which in turn changes the modal angle of incidence. The loss/bounce model calculates the loss of sound energy (dB) per bottom bounce over a given distance based on the change in angle of incidence. Evaluated using experimental data from NEST, the loss/bounce model provided accurate predictions of changes to transmission loss due to perturbations of the background sound speed field.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2009

Description: Mode filtering is most commonly implemented using the sampled mode shape or pseudoinverse algorithms. Buck et al placed these techniques in the context of a broader maximum a posteriori (MAP) framework. However, the MAP algorithm requires that the signal and noise statistics be known a priori. Adaptive array processing algorithms are candidates for improving performance without the need for a priori signal and noise statistics. A variant of the physically constrained, maximum likelihood (PCML) algorithm is developed for mode filtering that achieves the same performance as the MAP mode filter yet does not need a priori knowledge of the signal and noise statistics. The central innovation of this adaptive mode filter is that the received signal's sample covariance matrix, as estimated by the algorithm, is constrained to be that which can be physically realized given a modal propagation model and an appropriate noise model. The first simulation presented in this thesis models the acoustic pressure field as a complex Gaussian random vector and compares the performance of the pseudoinverse, reduced rank pseudoinverse, sampled mode shape, PCML minimum power distortionless response (MPDR), PCML-MAP, and MAP mode filters. The PCML-MAP filter performs as well as the MAP filter without the need for a priori data statistics. The PCML-MPDR filter performs nearly as well as the MAP filter as well, and avoids a sawtooth pattern that occurs with the reduced rank pseudoinverse filter. The second simulation presented models the underwater environment and broadband communication setup of the Shallow Water 2006 (SW06) experiment. Data processing results are presented from the Shallow Water 2006 experiment, showing the reduced sensitivity of the PCML-MPDR filter to white noise compared with the reduced rank pseudoinverse filter. Lastly, a linear, decision-directed, RLS equalizer is used to combine the response of several modes and its performance is compared with an equalizer applied directly to the data received on each hydrophone.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2009

Description: Marine mammal whistle calls present an attractive medium for covert underwater communications. High quality models of the whistle calls are needed in order to synthesize natural-sounding whistles with embedded information. Since the whistle calls are composed of frequency modulated harmonic tones, they are best modeled as a weighted superposition of harmonically related sinusoids. Previous research with bottlenose dolphin whistle calls has produced synthetic whistles that sound too “clean” for use in a covert communications system. Due to the sensitivity of the human auditory system, watermarking schemes that slightly modify the fundamental frequency contour have good potential for producing natural-sounding whistles embedded with retrievable watermarks. Structured total least squares is used with linear prediction analysis to track the time-varying fundamental frequency and harmonic amplitude contours throughout a whistle call. Simulation and experimental results demonstrate the capability to accurately model bottlenose dolphin whistle calls and retrieve embedded information from watermarked synthetic whistle calls. Different fundamental frequency watermarking schemes are proposed based on their ability to produce natural sounding synthetic whistles and yield suitable watermark detection and retrieval.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2002

Description: Sound propagation in shallow water is highly dependent on the interaction of the sound field with the bottom. In order to fully understand this problem, it is necessary to obtain reliable estimates of bottom geoacoustic properties that can be used in acoustic propagation codes. In this thesis, perturbative inversion methods and exact inverse methods are discussed as a means for inferring geoacoustic properties of the bottom. For each of these methods, the input data to the inversion is the horizontal wavenumber spectrum of a point-source acoustic field. The main thrust of the thesis work concerns extracting horizontal wavenumber content for fully three-dimensionally varying waveguide environments. In this context, a high-resolution autoregressive (AR) spectral estimator was applied to determine wavenumber content for short aperture data. As part of this work, the AR estimator was examined for its ability to detect discrete wavenumbers in the presence of noise and also to resolve closely spaced wavenumbers for short aperture data. As part of a geoacoustic inversion workshop, the estimator was applied to extract horizontal wavenumber content for synthetic pressure field data with range-varying geoacoustic properties in the sediment. The resulting wavenumber content was used as input data to a perturbative inverse algorithm to determine the sound speed profile in the sediment. It was shown using the high-resolution wavenumber estimator that both the shape and location of the range-variability in the sediment could be determined. The estimator was also applied to determine wavenumbers for synthetic data where the water column sound speed contained temporal variations due to the presence of internal waves. It was shown that reliable estimates of horizontal wavenumbers could be obtained that are consistent with the boundary conditions of the waveguide. The Modal Mapping Experiment (MOMAX), an experimental method for measuring the full spatial variability of a propagating sound field and its corresponding modal content in two-dimensions, is also discussed. The AR estimator is applied to extract modal content from the real data and interpreted with respect to source/receiver motion and geometry. For a moving source, it is shown that the wavenumber content is Doppler shifted. A method is then described that allows the direct measure of modal group velocities from Doppler shifted wavenumber spectra. Finally, numerical studies are presented addressing the practical issues associated with using MOMAX type data in the exact inversion method of Gelfand-Levitan.

Description: I am especially grateful to ONR for providing the funding for me to do this work.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2009

Description: My thesis covers two general circulation problems that involve the stability of largescale oceanic flows and the importance of non-local effects. The first problem examines the stability of meridional boundary currents, which are found on both sides of most ocean basins because of the presence of continents. A linear stability analysis of a meridional boundary current on the beta-plane is performed using a quasi-geostrophic model in order to determine the existence of radiating instabilities, a type of instability that propagates energy away from its origin region by exciting Rossby waves and can thus act as a source of eddy energy for the ocean interior. It is found that radiating instabilities are commonly found in both eastern and western boundary currents. However, there are some significant differences that make eastern boundary currents more interesting from a radiation point of view. They possess a larger number of radiating modes, characterized by horizontal wavenumbers which would make them appear like zonal jets as they propagate into the ocean interior. The second problem examines the circulation in a nonlinear thermally-forced two-layer quasi-geostrophic ocean. The only driving force for the circulation in the model is a cross-isopycnal flux parameterized as interface relaxation. This forcing is similar to the radiative damping used commonly in atmospheric models, except that it is applied to the ocean circulation in a closed basin and is meant to represent the large-scale thermal forcing acting on the oceans. It is found that in the strongly nonlinear regime a substantial, not directly thermally-driven barotropic circulation is generated. Its variability in the limit of weak bottom drag is dominated by high-frequency barotropic basin modes. It is demonstrated that the excitation of basin normal modes has significant consequences for the mean state of the system and its variability, conclusions that are likely to apply for any other system whose variability is dominated by basin modes, no matter the forcing. A linear stability analysis performed on a wind- and a thermally-forced double-gyre circulation reveals that under certain conditions the basin modes can arise from local instabilities of the flow.

Description: I was supported through a graduate research assistantship from the National Science Foundation Grant OCE-0423975 and the Woods Hole Oceanographic Institution Academic Programs Office.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution May 1984

Description: Part one of this thesis discusses the structure of the thermocline and the current pattern within a two-layer model. The corresponding flow field is explored as the amount of water in the upper layer is gradually reduced (or as the wind stress is gradually increased). In the model, when the amount of water in the upper layer is less than a first critical value, the lower layer outcrops near the middle of the western boundary. A dynamically consistent picture includes a whole loop of boundary currents, which surround the outcropping zone completely and have quite different structures. In addition to the boundary currents found in previous models, there is an isolated western boundary current (i.e. bounded on one side by the wall and on the other by a streamline along which the upper layer thickness vanishes), an internal boundary current and possibly isolated northern/southern boundary currents. Within the limitations of the two-layer model, the isolated western boundary current appears to represent the Labrador Current while the internal boundary current may represent the North Atlantic Current. A first baroclinic mode of water mass exchange occurs across the ZWCL (zero-wind-curl-line). When the amount of water in the upper layer is less than a second critical value, the upper layer separates from the eastern wall and becomes a warm water pool in the south-west corner of the basin. Under this warm water pool is the ventilated lower layer. The sea surface density distribution is not specified; it is determined from a consistent dynamical and mass balance. Implicit in this model is the assumption that advection dominates in the mixed layer. The subtropical gyre and the subpolar gyre combine asymmetrically with respect to the ZWCL. Chapter I discusses the case when the lower layer depth is infinite. Chapter II discusses the case when the lower layer depth is finite. In the Addendum the climatological meaning of this two-layer model is discussed. Part two of this thesis concerns the use of a continuously stratified model to represent the thermocline and current structures in subtropical/subpolar basins. The ideal fluid thermocline equation system Is a nonlinear, non-strict hyperbolic system. In an Addendum to Chapter III the mathematical properties of this equation system are studied and a proper way of formulating boundary value problems is discussed. Although the equations are not of standard type, so that no firm conclusions about the existence and uniqueness of solutions have been drawn, some possible approaches to properly posed boundary value problem are suggested. Chapter III presents some simple numerical solutions of the ideal fluid thermocline equation for a subtropical gyre and a subtropical/subpolar basin using one of these approaches. Our model predicts the continuous three dimensional thermocline and current structures in a continuously stratified wind-driven ocean. The upper surface density and Ekman pumping velocity are specified as input data; in addition, the functional form of the potential vorticity is specified. The present model emphasizes the idea that the ideal fluid thermocline model is incomplete. The potential vorticity distribution can not be determined within this idealized model. This suggests that the diffusion and upwelling/downwelling within the western boundary current and the outcropping zone in the north-west corner are important parts of the entire circulation system.

Description: This work was supported by NSF Grant 80-19260-0CE.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2009.

Description: This thesis examines the nature of eddy-mean flow interactions in western boundary current jets and recirculation gyre dynamics from both theoretical and observational perspectives. It includes theoretical studies of eddy-mean flow interactions in idealized configurations relevant to western boundary current jet systems, namely (i) a study of the mechanism by which eddies generated from a localized forcing drive mean recirculation gyres through the process of nonlinear rectification; and (ii) a study of the role of eddies in the downstream evolution of a baroclinic jet subject to mixed instabilities. It also includes an observational analysis to characterize eddy-mean flow interactions in the Kuroshio Extension using data from the downstream location of maximum eddy kinetic energy in the jet. New insights are presented into a rectification mechanism by which eddies drive the recirculation gyres observed in western boundary current systems. Via this mechanism, eddies drive the recirculations by an up-gradient eddy potential vorticity flux inside a localized region of eddy activity. The effectiveness of the process depends on the properties of the energy radiation from the region, which in turn depends on the population of waves excited. In the zonally-evolving western boundary current jet, eddies also act to stabilize the unstable jet through down-gradient potential vorticity fluxes. In this configuration, the role of eddies depends critically on their downstream location relative to where the unstable time-mean jet first becomes stabilized by the eddy activity. The zonal advection of eddy activity from upstream of this location is fundamental to the mechanism permitting the eddies to drive the mean flows. Observational results are presented that provide the first clear evidence of a northern recirculation gyre in the Kuroshio Extension, as well as support for the hypothesis that the recirculations are, at least partially, eddy-driven. Support for the idealized studies’ relevance to the oceanic regime is provided both by indications that various model simplifications are appropriate to the observed system, as well as by demonstrated consistencies between model predictions and observational results in the downstream development of time-mean and eddy properties.

Description: Funding was for this research and my education was provided by the MIT Presidential Fellowship and NSF grants OCE-0220161 and OCE-0825550. The financial assistance of the Houghton Fund, the MIT Student Assistance Fund, and WHOI Academic Programs is also gratefully acknowledged.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2007

Description: This thesis covers a comprehensive analysis of long-range, deep-ocean, low-frequency, sound propagation experimental results obtained from the North Pacific Ocean. The statistics of acoustic fields after propagation through internal-wave-induced sound-speed fluctuations are explored experimentally and theoretically. The thesis starts with the investigation of the North Pacific Acoustic Laboratory 98-99 data by exploring the space-time scales of ocean sound speed variability and the contributions from different frequency bands. The validity of the Garret & Munk internal-wave model is checked in the upper ocean of the eastern North Pacific. All these results impose hard bounds on the strength and characteristic scales of sound speed fluctuations one might expect in this region of the North Pacific for both internal-wave band fluctuations and mesoscale band fluctuations. The thesis then presents a detailed analysis of the low frequency, broadband sound arrivals obtained in the North Pacific Ocean. The observed acoustic variability is compared with acoustic predictions based on the weak fluctuation theory of Rytov, and direct parabolic equation Monte Carlo simulations. The comparisons show that a resonance condition exists between the local acoustic ray and the internal wave field such that only the internal-waves whose crests are parallel to the local ray path will contribute to acoustic scattering: This effect leads to an important filtering of the acoustic spectra relative to the internal-wave spectra. We believe that this is the first observational evidence for the acoustic ray and internal wave resonance. Finally, the thesis examined the evolution with distance, of the acoustic arrival pattern of the off-axis sound source transmissions in the Long-range Ocean Acoustic Propagation EXperiment. The observations of mean intensity time-fronts are compared to the deterministic ray, parabolic equation (with/without internal waves) and (one-way coupled) normal mode calculations. It is found the diffraction effect is dominant in the shorter-range transmission. In the longer range, the (internal wave) scattering effect smears the energy in both the spatial and temporal scales and thus has a dominant role in the finale region.

Description: The funding that made this research possible came from the Office of Naval Research, and the WHOI Academic Programs Office.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2008

Description: The subtidal circulation of the southeast Greenland shelf is described using a set of highresolution hydrographic and velocity transects occupied in summer 2004. The main feature present is the East Greenland Coastal Current (EGCC), a low-salinity, highvelocity jet with a wedge-shaped hydrographic structure characteristic of other surface buoyancy-driven currents. The EGCC was observed along the entire Greenland shelf south of Denmark Strait, while the transect north of the strait showed only a weak shelf flow. This observation, combined with evidence from chemical tracer measurements that imply the EGCC contains a significant Pacific Water signal, suggests that the EGCC is an inner branch of the polar-origin East Greenland Current (EGC). A set of idealized laboratory experiments on the interaction of a buoyant current with a submarine canyon also supported this hypothesis, showing that for the observed range of oceanic parameters, a buoyant current such as the EGC could exhibit both flow across the canyon mouth or into the canyon itself, setting the stage for EGCC formation. Repeat sections occupied at Cape Farewell between 1997 and 2004 show that the alongshelf wind stress can also have a strong influence on the structure and strength of the EGCC and EGC on timescales of 2-3 days. Accounting for the wind-induced effects, the volume transport of the combined EGC/EGCC system is found to be roughly constant (~2 Sv) over the study domain, from 68°N to Cape Farewell near 60°N. The corresponding freshwater transport increases by roughly 60% over this distance (59 to 96 mSv, referenced to a salinity of 34.8). This trend is explained by constructing a simple freshwater budget of the EGCC/EGC system that accounts for meltwater runoff, melting sea-ice and icebergs, and net precipitation minus evaporation. Variability on interannual timescales is examined by calculating the Pacific Water content in the EGC/EGCC from 1984-2004 in the vicinity of Denmark Strait. The PW content is found to correlate significantly with the Arctic Oscillation index, lagged by 9 years, suggesting that the Arctic Ocean circulation patterns bring varying amounts of Pacific Water to the North Atlantic via the EGC/EGCC.

Description: Funding for the cruise and analysis was provided by National Science Foundation grant OCE-0450658, which along with NSF grant OCE- 0095427 provided funds for my tuition and stipend as well.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2003

Description: A novel distributed underwater acoustic networking (UAN) protocol suitable for ad-hoc deployments of both stationary and mobile nodes dispersed across a relatively wide coverage area is presented. Nodes are dynamically clustered in a distributed manner based on the estimated position of one-hop neighbor nodes within a shallow water environment. The spatial dynamic cellular clustering scheme allows scalable communication resource allocation and channel reuse similar in design to land-based cellular architectures, except devoid of the need for a centralized controlling infrastructure. Simulation results demonstrate that relatively high degrees of interference immunity, network connectivity, and network stability can be achieved despite the severe limitations of the underwater acoustic channel.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution June 2002

Description: This thesis investigates the complexities of acoustic scattering by finite bodies in general and by fish in particular through the development of an advanced acoustic scattering model and detailed laboratory acoustic measurements. A general acoustic scattering model is developed that is accurate and numerically effcient for a wide range of frequencies, angles of orientation, irregular axisymmetric shapes and boundary conditions. The model presented is an extension of a two-dimensional conformal mapping approach to scattering by irregular, finite-length bodies of revolution. An extensive series of broadband acoustic backscattering measurements has been conducted involving alewife fish (Alosa pseudoharengus), which are morphologically similar to the Atlantic herring (Clupea harengus). A greater-than-octave bandwidth (40-95 kHz), shaped, linearly swept, frequency modulated signal was used to insonify live, adult alewife that were tethered while being rotated in 1-degree increments over all angles of orientation in two planes of rotation (lateral and dorsal/ventral). Spectral analysis correlates frequency dependencies to morphology and orientation. Pulse compression processing temporally resolves multiple returns from each individual which show good correlation with size and orientation, and demonstrate that there exists more than one significant scattering feature in the animaL. Imaging technologies used to exactly measure the morphology of the scattering features of fish include very highresolution Phase Contrast X-rays (PCX) and Computerized Tomography (CT) scans, which are used for morphological evaluation and incorporation into the scattering modeL. Studies such as this one, which combine scattering models with high-resolution morphological information and high-quality laboratory data, are crucial to the quantitative use of acoustics in the ocean.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution April 1983

Description: Ocean acoustic tomography was proposed in 1978 by Munk and Wunsch as a possible technique for monitoring the evolution of temperature, density, and current fields over large regions. In 1981, the Ocean Tomography Group deployed four 224 Hz acoustic sources and five receivers in an array which fit within a box 300 km. on a side centered on 26°N, 70°W (southwest of Bermuda). The experiment was intended both to demonstrate the practicality of tomography as an observation tool and to extend the understanding of mesoscale evolution in the low-energy region far from the strong Gulf Stream recirculation. The propagation of 224 Hz sound energy in the ocean can be described as a set of rays traveling from source to receiver, with each ray taking a different path through the ocean in a vertical plane connecting the source and receiver. The sources transmitted a phase-coded signal which was processed at the receiver to produce a pulse at the time of arrival of the signal. Rays can be distinguished by their different pulse travel times, and these travel times change in response to variations in sound speed and current in the ocean through which the rays passed. In order to reconstruct the ocean variations from the observed travel time changes, it is necessary to specify models for both the variations and their effect on the travel times. The dependence of travel time on the oceanic sound speed and current fields can be calculated using ray paths traced by computer. The vertical structure of the sound speed and current fields in the ocean were modelled as a combination of Empirical Orthogonal Functions (EOFs) from MODE. The horizontal structure was continuous, but was constrained to have a gaussian covariance with a 100 km. e- folding scale. The resulting estimator closely resembles objective mapping as used in meteorology and physical oceanography. The tomographic system has at present only been used to estimate sound speed structure for comparison with the traditional measurements, especially the first two NOAA CTD surveys, but the method provides means for estimating density, temperature or velocity fields, and these will be produced in the future. The sound speed estimates made using the tomographic system match the traditional measurements to within the associated error bars, and there are several possibilities for improving the signal to noise ratio of the data. Given high-precision data, tomographic systems can resolve ocean structures at small scales, such as in the Gulf Stream, or at large scales, over entire ocean basins. Work is in progress to evaluate the usefulness of tomography as an observation tool in these applications.

Description: My support for the first 3 years came from an NSF graduate fellowship, and I was then supported as a research assistant by NSF Grant OCE-8017791.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1982

Description: Mean long-isobath drift of the order 5 cm/sec has been observed on several continental shelves, e.g. in the Middle Atlantic Bight and in the Weddell Sea. A theoretical model is developed to explore the driving mechanism of this mean circulation. In the model, the velocity field is decomposed into a depth-independent bottom geostrophic component and a thermohaline component relative to the bottom. The latter can be calculated from the density field, and the former is described by a parabolic equation which expresses the tendency-to balance vorticity between bottom stress curl and vortex stretching. The near-bottom flow field is studied both analytically and numerically under forcing by wind, deep ocean flow, and long-isobath density differences. Model solutions are derived for circulations over a shelf/slope topography driven by wind stress, wind stress curl, and deep ocean currents. The resulting flow patterns show strong dependence on the topography. Over the continental slope, large bottom depth variation suppresses the flow driven by local forcing and insulates the slope region from circulations on the shelf and in the deep-ocean. Geochemical observations on the continental shelf and slope support the argument that the flow on the upper slope below the thermocline is weak. Under the condition of a vertically homogeneous layer below the thermocline, near-bottom density advection is mainly caused by the bottom geostrophic velocity field. Using the parabolic vorticity equation together with a density equation, circulations driven by coastal buoyancy flux and surface cooling are investigated. In the mid-shelf region, away from the coast and the shelf break, the density field is governed by Burgers' equation, which shows longshore self-advection of density perturbations and the formation of front with strong density gradient in the longshore direction. A dense water blob moves in the direction of Kelvin wave propagation. The direction is reversed for the movement of a light water blob. In the near-shore region, the light river water bottom is also self-advected in the direction of Kelvin wave propagation. For a heavy density anomaly at the coast, the initial movement is offshore, and the accumulation of dense water in the mid-shelf region leads to long-isobath propagation of density perturbations, similar to the case of a dense water blob. This theory sheds light on the bottom water movements in the Adriatic Sea, the Antarctic Continent, and the Middle Atlantic Bight. The model solutions are applied to the flow on the western North Atlantic shelf. Southwestward flow is produced near the coast by the self-advection of river water in winter and spring. The southwestward long-isobath propagation of thermal fronts caused by winter cooling contributes significantly to the mean circulation over the mid-shelf. It is suggested that density-driven current is an important component of the near-bottom mean circulation in the Middle Atlantic Bight in spring and summer.

Description: This work was supported by the Department of Energy through contract entitled Coastal-Shelf Transport and Diffusion.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute and the Woods Hole Oceanographic Institution August 1975

Description: This thesis is made of two separate, but interrelated parts. In Part I the instability of a baroclinic Rossby wave in a two-layer ocean of inviscid fluid without topography, is investigated and its results are applied in the ocean. The velocity field of the basic state (the wave) is characterized by significant horizontal and vertical shears, non-zonal currents, and unsteadiness due to its westward propagation. This configuration is more relevant to the ocean than are the steady, zonal 'meteorological' flows, which dominate the literature of baroclinic instability. Truncated Fourier series are used in perturbation analyses. The wave is found to be unstable for a wide range of the wavelength; growing perturbations draw their energy from kinetic or potential energy of the wave depending upon whether the wavelength, 2πL, is much smaller or larger than 2πLρ, respectively, where Lρ is the internal radius of deformation. When the shears are comparable dynamically, L~Lρ , the balance between the two energy transfer processes is very sensitive to the ratios L/Lρ and U/C as well, where U is a typical current speed, and C a typical phase speed of the wave. For L = Lρ they are augmenting if U 〈 C, yet they detract from each other if U 〉 C. The beta-effect tends to stabilize the flow, but perturbations dominated by a zonal velocity can grow irrespective of the beta-effect. It is necessary that growing perturbations are comprised of both barotropic and baroclinic modes vertically. The scale of the fastest growing perturbation is significantly larger than L for barotropically controlled flows (L 〈 Lρ ), reduces to the wave scale L for a mixed kind (L ~ Lρ ) and is fixed slightly larger than Lρ for baroclinically controlled flows (L 〉 Lρ ). Increasing supply of potential energy causes the normalized growth rate, αL/U, to increase monotonically as L → Lρ from below. As L increases beyond Lρ, the growth rate αLρ /U shows a slight increase, but soon approaches an asymptotic value. In a geophysical eddy field like the ocean this model shows possible pumping of energy into the radius of deformation (~ 40 km rational scale, or 250 km wavelength) from both smaller and larger scales through nonlinear interactions, which occur without interference from the beta-effect. The e-folding time scale is about 24 days if U = 5 cm/sec and L = 90 km. Also it is strongly suggested that, given the observed distribution of energy versus length scale, eddy-eddy interactions are more vigorous than eddy-mean interaction, away from intènse currents like the Gulf Stream. The flux of energy toward the deformation scale, and the interaction of barotropic and baroclinic modes, occur also in fully turbulent 'computer' oceans, and these calculations provide a theoretical basis for source of these experimental cascades. In Part II an available potential energy (APE) is defined in terms appropriate to a limited area synoptic density map (e.g., the 'MODE-I' data) and then in terms appropriate to time-series of hydrographic station at a single geographic location (e. g., the 'Panulirus' data). Instantaneously the APE shows highly variable spatial structure, horizontally as well as vertically, but the vertical profile of the average APE from 19 stations resembles the profile of vertical gradient of the reference stratification. The eddy APE takes values very similar to those of the average kinetic energy density at 500 m, 1500 m and 3000 m depth in the MODE area. In and above the thermocline the APE has roughly the same level in the MODE area (centered at 28°N, 69° 40'W) as at the Panulirus station (32° 10'N, 64° 30'W), yet in the deep water there is significantly more APE at the Panulirus station. This may in part indicate an island effect near Bermuda.

Description: This research has been supported by the National Science Foundation grant IDO 73-09737, formerly GX-36342.