The author has employed ocean acoustic tomography over the past 30 years to examine problems in physical oceanography, often with surprising results. Tomography offers an accurate measurement of current or sound speed (temperature) averaged over 100s to 1000s of kilometers. It has been used to test the fundamental equation for the speed of sound in seawater and to show that the normal state of the ocean has smooth, stable, well-behaved characteristics of acoustic propagation. This latter property has been challenging for ocean models to reproduce. Tomography measures barotropic tidal currents with remarkable accuracy, and it was used to discover that low-mode internal tides radiate far into the ocean’s interior, while retaining a surprising coherence. Tomographic measurements of large-scale temperature are complementary to point measurements by hydrography; the present observing system, relying principally on altimetry and Argo floats, has little skill in predicting the available tomographic data. The addition of tomographic data to the system would substantially reduce the uncertainty of ocean state estimates. The information content of integral acoustic observations is best exploited as a constraint on ocean models by data assimilation. The applications of tomography for measuring large-scale barotropic current, relative vorticity, and temperature have been under-exploited. Dedicated research programs, supporting novel acoustical oceanography, associated instrumentation development, and multidisciplinary research would further the applications of this observational approach.

Operation WIGWAM was a test of a 30 kt nuclear depth charge conducted in deep water 500 miles southwest of San Diego on 14 May 1955. Its primary purpose was to determine the effectiveness of that device as an antisubmarine weapon. The acoustic pulse from the test, initially an intense shockwave, radiated throughout the North and South Pacific Oceans. Acoustic reflections from topographic features were recorded for several hours after the explosion by SOund Fixing And Ranging (SOFAR) hydrophones at Point Sur, California, and Kaneohe, Hawaii. Sheehy and Halley (1957) identified peaks of the recorded coda with reflections from specific topographic features at great distances (e.g., the Hawaiian Islands, French Polynesia, or Fiji). With modern data for seafloor topography and ocean sound speed, these coda were computed with surprising accuracy using simple geodesic rays reflected from islands and seamounts. The intensity variations of the coda are mostly determined by simple ray geometry, together with modest attenuation. Coda peaks are often obtained from rays arriving simultaneously from multiple, but disparate, topographic features.

The dramatic reduction of sea ice in the Arctic Ocean will increase human activities in the coming years. This activity will be driven by increased demand for energy and the marine resources of an Arctic Ocean accessible to ships. Oil and gas exploration, fisheries, mineral extraction, marine transportation, research and development, tourism, and search and rescue will increase the pressure on the vulnerable Arctic environment. Technologies that allow synoptic in situ observations year-round are needed to monitor and forecast changes in the Arctic atmosphere-ice-ocean system at daily, seasonal, annual, and decadal scales. These data can inform and enable both sustainable development and enforcement of international Arctic agreements and treaties, while protecting this critical environment. In this paper, we discuss multipurpose acoustic networks, including subsea cable components, in the Arctic. These networks provide communication, power, underwater and under-ice navigation, passive monitoring of ambient sound (ice, seismic, biologic, and anthropogenic), and acoustic remote sensing (tomography and thermometry), supporting and complementing data collection from platforms, moorings, and vehicles. We support the development and implementation of regional to basin-wide acoustic networks as an integral component of a multidisciplinary in situ Arctic Ocean observatory.

The analysis of signals for acoustic tomography sent between a source and a receiver most often uses the unrefracted geodesic path, an approximation that is justified from theoretical considerations, relying on estimates of horizontal gradients of sound speed, or on simple theoretical models. To quantify the effects of horizontal refraction caused by a realistic ocean environment, horizontal refractions of long-range signals were computed using global ocean state estimates for 2004 from the Estimating the Circulation and Climate of the Ocean (ECCO2) project. Basin-scale paths in the eastern North Pacific Ocean and regional-scale paths in the Philippine Sea were used as examples. At O(5 Mm) basin scales, refracted geodesic and geodesic paths differed by only about 5%u2009km. Gyre-scale features had the greatest refractive influence, but the precise refractive effects depended on the path geometry with respect to oceanographic features. Refraction decreased travel times by 5%u201310%u2009ms and changed azimuthal angles by about 0.2°. At O(500%u2009km) regional scales, paths deviated from the geodesic by only 250%u2009m, and travel times deviated by less than 0.5%u2009ms. Such effects are of little consequence in the analysis of tomographic data. Refraction details depend only slightly on mode number and frequency.

On 21 March 1960, sounds from three 300-lb depth charges deployed at 5.5-min intervals off Perth, Australia were recorded by the SOFAR station at Bermuda. The recorded travel time of these signals, about 13,375 s, is a historical measure of the ocean temperature averaged across several ocean basins. The 1960 travel time measurement has about 3-s precision. High-resolution global ocean state estimates for 2004 from the "Estimating the Circulation and Climate of the Ocean, Phase II" (ECCO2) project were combined with ray tracing to determine the paths followed by the acoustic signals. The acoustic paths are refracted geodesics that are slightly deflected by either small-scale topographic features in the Southern Ocean or the coast of Brazil. The refractive influences of intense, small-scale oceanographic features, such as Agulhas Rings or eddies in the Antarctic Circumpolar Current, greatly reduce the necessary topographic deflection and cause the acoustic paths to meander in time. The ECCO2 ocean state estimates, which are constrained by model dynamics and available data, were used to compute present-day travel times. Measured and computed arrival coda were in good agreement. Based on recent estimates of warming of the upper ocean, the travel-time change over the past half-century was nominally expected to be about –9 s, but little difference between measured (1960) and computed (2004) travel times was found. Taking into account uncertainties in the 1960 measurements, the 2004 ocean state estimates, and other approximations, the ocean temperature averaged along the sound channel axis over the antipodal paths has warmed at a rate less than about 4.6 m°C yr^-1 (95% confidence). At this time, the estimated uncertainties are comparable in size to the expected warming signal, however.

As an aid to understanding long-range acoustic propagation in the Philippine Sea, statistical and phenomenological descriptions of sound-speed variations were developed. Two moorings of oceanographic sensors located in the western Philippine Sea in the spring of 2009 were used to track constant potential-density surfaces (isopycnals) and constant potential-temperature surfaces (isotherms) in the depth range 120–2000 m. The vertical displacements of these surfaces are used to estimate sound-speed fluctuations from internal waves, while temperature/salinity variability along isopycnals are used to estimate sound-speed fluctuations from intrusive structure often termed spice. Frequency spectra and vertical covariance functions are used to describe the space-time scales of the displacements and spiciness. Internal-wave contributions from diurnal and semi-diurnal internal tides and the diffuse internal-wave field [related to the Garrett–Munk (GM) spectrum] are found to dominate the sound-speed variability. Spice fluctuations are weak in comparison. The internal wave and spice frequency spectra have similar form in the upper ocean but are markedly different below 170-m depth. Diffuse internal-wave mode spectra show a form similar to the GM model, while internal-tide mode spectra scale as mode number to the minus two power. Spice decorrelates rapidly with depth, with a typical correlation scale of tens of meters.

A series of experiments conducted in the Philippine Sea during 2009–2011 investigated deep-water acoustic propagation and ambient noise in this oceanographically and geologically complex region: (i) the 2009 North Pacific Acoustic Laboratory (NPAL) Pilot Study/Engineering Test, (ii) the 2010–2011 NPAL Philippine Sea Experiment, and (iii) the Ocean Bottom Seismometer Augmentation of the 2010–2011 NPAL Philippine Sea Experiment. The experimental goals included (a) understanding the impacts of fronts, eddies, and internal tides on acoustic propagation, (b) determining whether acoustic methods, together with other measurements and ocean modeling, can yield estimates of the time-evolving ocean state useful for making improved acoustic predictions, (c) improving our understanding of the physics of scattering by internal waves and spice, (d) characterizing the depth dependence and temporal variability of ambient noise, and (e) understanding the relationship between the acoustic field in the water column and the seismic field in the seafloor. In these experiments, moored and ship-suspended low-frequency acoustic sources transmitted to a newly developed distributed vertical line array receiver capable of spanning the water column in the deep ocean. The acoustic transmissions and ambient noise were also recorded by a towed hydrophone array, by acoustic Seagliders, and by ocean bottom seismometers.

Receptions on three vertical hydrophone arrays from basin-scale acoustic transmissions in the North Pacific during 1996 and 1998 are used to test the time-mean sound-speed properties of the World Ocean Atlas 2005 (WOA05), of an eddying unconstrained simulation of the Parallel Ocean Program (POP), and of three data-constrained solutions provided by the estimating the circulation and climate of the ocean (ECCO) project: a solution based on an approximate Kalman filter from the Jet Propulsion Laboratory (ECCO-JPL), a solution based on the adjoint method from the Massachusetts Institute of Technology (ECCO-MIT), and an eddying solution based on a Green's function approach from ECCO, Phase II (ECCO2). Predictions for arrival patterns using annual average WOA05 fields match observations to within small travel time offsets (0.3–1.0 s). Predictions for arrival patterns from the models differ substantially from the measured arrival patterns, from the WOA05 climatology, and from each other, both in terms of travel time and in the structure of the arrival patterns. The acoustic arrival patterns are sensitive to the vertical gradients of sound speed that govern acoustic propagation. Basin-scale acoustic transmissions, therefore, provide stringent tests of the vertical temperature structure of ocean state estimates. This structure ultimately influences the mixing between the surface waters and the ocean interior. The relatively good agreement of the acoustic data with the more recent ECCO solutions indicates that numerical ocean models have reached a level of accuracy where the acoustic data can provide useful additional constraints for ocean state estimation.

Acoustic data acquired during the 1995-2006 Acoustic Thermometry of Ocean Climate (ATOC) program were used to test the accuracy of ocean state estimates of the North Pacific obtained by various means: simple forward integration of a model, objective analysis of hydrographic and altimeter data, and data assimilation using general circulation models. The comparisons of computed and measured time series stringently tested the accuracy of the state estimates. The differences were substantial, indicating that acoustic thermometry provides unique information about the large-scale temperature. On some acoustic paths, changes in temperature occurring over time scales of weeks with magnitudes comparable to the seasonal cycle were observed. Acoustic thermometry offers valuable constraints on the large-scale thermal variability for the ocean observing system. Acoustic tomography was accepted as part of the Ocean Observing System during the OceanObs'99 and '09 international workshops. Sources and receivers of acoustic thermometry can serve multiple purposes. Hydrophone arrays are used to study a wide range of human, biological, and geological activity. Acoustic sources can transmit signals that can be used to track drifting instrumentation. A modest number of active and passive acoustic instruments deployed worldwide can form a general purpose global acoustic observing network.

From April 2010 to April 2011, 6 moorings equipped with temperature (T), conductivity (C), and pressure (P) sensors along with ADCP's observed oceanic variability in support of concurrent acoustic measurements between the moorings. In addition, for the month of April in 2009, two moorings monitored ocean conditions for a pilot study. During these periods energetic internal waves and tides, as well as eddies were observed thus creating an inhomogeneous, anisotropic, and rapidly changing acoustical environment. Some moorings possessed high precision T, C, and P records capable of resolving intrusive structures sometimes termed spice. In this talk statistical and deterministic metrics will be used to characterize the various dynamical sources of sound speed variability that were observed.

A frequency-wavenumber tidal analysis for deriving internal-tide harmonic constants from TOPEX/Poseidon (T/P) measurements of sea-surface height (SSH) has been developed, taking advantage of the evident temporal and spatial coherence and the weak dissipation of internal tides. Previous analyses consisted of simple tidal analysis at individual points, which gave inconsistent harmonic constants at altimeter track crossover points. Such analyses have difficulty in distinguishing between the effects of interference, incoherence, and dissipation. The frequency-wavenumber analysis provides an objective way to interpolate the internal tides measured along altimetry tracks to any arbitrary point, while leveraging all available data for optimal tidal estimates. Tidal analysis of T/P data from 2000 to 2007 is used to predict in situ time series measured during the 2001-2002 Hawaiian Ocean mixing experiment (HOME), the 1987 reciprocal tomography experiment (RTE87), and the 1991 acoustic mid-ocean dynamics experiment (AMODE), demonstrating both the temporal coherence and the lack of incoherent elements to this wave propagation. It has been conjectured that significant energy would be lost from mode-1 internal tides as they cross the 28.9 N critical latitude of parametric subharmonic instability (PSI). No apparent change in amplitude at 28.9 N was detected by this analysis, however. Further, after correcting for changes in background stratification, the amplitude of the mode-1 internal tide was found to decrease by less than 20% over the 2000 km between the Hawaiian Ridge and 40 N. A significant fraction of the variability of internal waves, that component associated with mode-1 internal tides, appears to be predictable over most of the world's oceans, using harmonic constants derived from satellite altimetry.

Integrated ocean observing systems (IOOSs) originated during the early 1990s as a way to provide valuable information to society as return on decades of investment in oceanographic research. The archetypical Argo program (www.argo.ucsd.edu) was initiated following the landmark OceanObs'99 conference and reassessed during OceanObs'09. With the possibility of significant sustained funding for comprehensive oceanic observations, one hallmark of planning conferences has been their political overtones. Regional ocean observing systems (www.ioos.gov, www.usnfra.org, and www.nanoos.org), purposely distinct from NOAA, began to develop in the early 2000s. The OOSes are not research programs, but sustained monitoring and information%u2010providing services. The ocean observing initiative (OOI) developed out of the deep earth observing system (DEOS) idea of the early 2000s to deploy platforms over the world's oceans for geophysical research. With the advent of significant funding for such platforms, the geophysical thrust was thrust aside in favor of two components: the undersea cabled networks and a small set of "Pioneer" platforms for oceanographic research. There has been little representation of acoustical techniques in these programs. Successful implementation of acoustical components for the IOOS requires the sustained cooperation and encouragement of three factions: the oceanographic community, the national and state funding agencies, and the acoustics community.

The uniqueness and sustainability of acoustic measurements of basin-scale temperature were demonstrated by the the Acoustic Thermometry of Ocean Climate and North Pacific Acoustic Laboratory programs in the North Pacific over the decade 1996-2006. Tomography has a role to play in the global observing system as a measurement type that complements altimetry and profiling floats. Tomography is a subsurface measurement of temperature; salinity contributes negligibly. Acoustic travel times are inherently precise integral measurements of temperature; mesoscale variability is supressed. Tomographic measurements offer one of the few ways to sample the abyssal ocean. The North Atlantic and Arctic have been highlighted by international conferences as regions that would be suitable for implementing tomographic arrays. Ocean basins can be measured by tomography using a few acoustic sources and receivers, employing platforms of opportunity such as existing or planned components of the ocean observing system. Extensive acoustic sampling can be achieved by sharing resources, while minimizing long-term operation and maintenance costs. Because of the integral nature of the data, tomography is best employed in conjunction with numerical ocean models and data assimilation.

Over the past decade, the oceanographic community has been constructing integrated multidisciplinary observational systems to serve the immediate and long-term needs of the society (www.ioos.gov). The overarching goal is to provide practical products and information to society as a return on its long-term investment in oceanographic research. Demand is increasing for information about the state of our oceans to address a myriad of issues ranging from climate variability to fisheries management to public education. Customers for ocean observing systems range from government agencies to commercial shipping companies to the science projects of high school students. Two developing "observing systems" in the Pacific Northwest are the Northwest Association of Networked Ocean Observing Systems (www.nanoos.org) and the Neptune Undersea Cable Network (www.interactiveoceans.washington.edu). The oceans are largely transparent to sound; hence, oceanographic, biological, and signal-processing acoustic techniques are primary tools for ocean observation and engineering. The opportunities and value of acoustical observations and techniques within these systems are boundless, yet incorporation of these techniques has been opportunistic and ad hoc. Coordination of the acoustical applications is essential. Organizations advocating acoustical observations face enormous challenges of planning, implementation, and data management to bring acoustical tools to fruition for ocean observing systems.

In 1960, the sound signals from a sequence of 300-lb TNT shots deployed off Perth, Australia were recorded at Bermuda. The acoustic propagation of these signals are examined using ECCO2 1/6 degree ocean state estimates [1992–2006; www.ecco2.org] and Smith–Sandwell 1' global topography. Only the first acoustic mode need be considered. Intense, small-scale features (e.g., Agulhas rings, the Antarctic circumpolar current) greatly influence the acoustic paths, giving them a scintillating and time-dependent nature. Previous analyzes found, in the absence of the influence of the sea floor, Bermuda to be in the acoustic shadow of the African continent.

The oceanic features of ECCO2 are sharper, hence more refractive, than the smoothed ocean atlases previously employed, but the additional refraction still does not give "direct" arrivals. The "direct" arrivals miss Bermuda by only 135 plus/minus 52 km, however. Scattering from Kerguelen and the Crozets, and reflection from the eastern tip of South America are sufficient to account for the 1960 measured arrivals. Intensity considerations show that because Bermuda was shadowed, the recorded signal levels of the explosive shots (15 dB RE noise) were greatly reduced compared to theoretically expected signal levels (30 dB RE noise).

Tidal analysis of satellite altimeter data has demonstrated that mode-1 internal tides near Hawaii have considerable predictability. Temporally incoherent contributions to mode-1 internal tides appear to be minimal, and very weak attenuation is evident. The internal-tide modes are associated functions of displacement, temperature, sound speed, current, etc. Sea-surface height is determined by the displacement mode. Whether derived from thermistors, altimetry, or acoustic tomography, the mode amplitude implies specific profiles of temperature, current, etc. The tidal analysis of altimeter data for the region of the Philippine Sea will be discussed, leading to predictions of in situ variability. These predictions will be compared to thermistor and tomographic data acquired in the Philippine Sea during the ONR-sponsored, PhilSea09 test experiment in March 2009.

These data consist of month-long measurements of temperature acquired by thermistors on two moorings, affording excellent resolution of the internal-tide modes. One of these moorings hosted a 225–325-Hz swept-frequency source that transmitted tomographic signals recorded on the other mooring by a water-column-spanning, distributed vertical line array. Strong, coherent internal-tide signals, particularly mode-1, are apparent in these data. A comparison of the altimetry and in situ results tests the predictability of low-mode internal tides in the Philippine Sea basin.

Many mid-latitude deep ocean environments are characterized by strongly range-dependent sound speed structure associated with mesoscale variability, including internal tides. Often much of this variability is describable as a combination of barotropic and lowest baroclinic mode variability. There are reasons to anticipate that ocean structure of this type leads only to a relatively benign range-dependent modulation of underwater sound fields. These ideas will be explained and tested using ECCO2 state estimates in the Philippine Sea.

The Philippine Sea is a dynamic ocean basin with complex multi-scale sound speed structure. Therefore the PhilSea09 and PhilSea10 experiments have put significant resources toward quantifying the space-time scales of this sound speed variability, so that the acoustic transmission data can be properly interpreted. In the PhilSea09 pilot study, two moorings equipped with temperature (T), conductivity (C), and pressure sensors, along with upper ocean ADCP, monitored ocean variability for a month in the Spring.

The measurements reveal an energetic and nonlinear mixed diurnal-semidiurnal internal tide, a diffuse Garrett–Munk (GM) type internal wave field at or above the reference GM energy level, and a strong eddy field. One mooring, which was equipped with pumped sensors for enhanced salinity (S) resolution, was able to accurately quantify T and S variability along isopycnals (spice). The spice contribution to sound speed fluctuation is strong near the mixed layer but is significantly weaker than the other contributions in the main thermocline. Frequency spectra as well as vertical covariance functions will be presented to quantify the temporal and vertical spatial scales of the observed fluctuations.

Most of the M2 internal tide energy generated at the Hawaiian Ridge radiates away in modes 1 and 2, but direct observation of these propagating waves is complicated by the complexity of the bathymetry at the generation region and by the presence of interference patterns.

Observations from satellite altimetry, a tomographic array, and the R/P FLIP taken during the Farfield Program of the Hawaiian Ocean Mixing Experiment (HOME) are found to be in good agreement with the output of a high-resolution primitive equation model, simulating the generation and propagation of internal tides. The model shows that different modes are generated with different amplitudes along complex topography. Multiple sources produce internal tides that sum constructively and destructively as they propagate. The major generation sites can be identified using a simplified 2D idealized knife-edge ridge model. Four line sources located on the Hawaiian Ridge reproduce the interference pattern of sea surface height and energy flux density fields from the numerical model for modes 1 and 2. Waves from multiple sources and their interference pattern have to be taken into account to correctly interpret in situ observations and satellite altimetry.

Over the decade 1996–2006, acoustic sources located off central California (1996–1999) and north of Kauai (1997–1999, 2002–2006) transmitted to receivers distributed throughout the northeast and north central Pacific. The acoustic travel times are inherently spatially integrating, which suppresses mesoscale variability and provides a precise measure of ray-averaged temperature. Daily average travel times at 4-day intervals provide excellent temporal resolution of the large-scale thermal field. The interannual, seasonal, and shorter-period variability is large, with substantial changes sometimes occurring in only a few weeks. Linear trends estimated over the decade are small compared to the interannual variability and inconsistent from path to path, with some acoustic paths warming slightly and others cooling slightly.

The measured travel times are compared with travel times derived from four independent estimates of the North Pacific: (1) climatology, as represented by the World Ocean Atlas 2005 (WOA05); (2) objective analysis of the upper-ocean temperature field derived from satellite altimetry and in situ profiles; (3) an analysis provided by the Estimating the Circulation and Climate of the Ocean project, as implemented at the Jet Propulsion Laboratory (JPL-ECCO); and (4) simulation results from a high-resolution configuration of the Parallel Ocean Program (POP) model. The acoustic data show that WOA05 is a better estimate of the time mean hydrography than either the JPL-ECCO or the POP estimates, both of which proved incapable of reproducing the observed acoustic arrival patterns. The comparisons of time series provide a stringent test of the large-scale temperature variability in the models. The differences are sometimes substantial, indicating that acoustic thermometry data can provide significant additional constraints for numerical ocean models.

In 1960, sound signals traveling from Perth, Australia were recorded at Bermuda. Previous work focused on the path traveled by the sound [Munk et al., JPO, 1876–1898 (1988)]. Calculation of the horizontal refraction of sound, across the Southern Ocean in particular, gave the perplexing result that Bermuda was in the shadow of Africa. Heaney et al. [J. Acoust. Soc. Am., 2586–2594 (1991)] used low-resolution atlases for global sound speed and bathymetry to obtain two viable acoustic paths between Perth and Bermuda, both influenced by bathymetry. From a modern perspective, however, the explanation of Heaney et al. is unconvincing [Dushaw, GRL (2008)].

High-resolution ocean models put the Perth-to-Bermuda acoustic problem into a new light. These models suggest that intense, small-scale features, e.g., Agulhas rings near the Cape of Good Hope, would greatly influence the acoustic paths. The antipodal travel time, 13 382 s, is a measure of the ocean temperature in 1960. If the acoustic propagation issues can be fully understood, data-assimilating ocean general circulation models might be used to calculate a present-day travel time. The travel-time change over the past half-century, expected to be about 10 s based on nominal estimates of ocean warming, is a measure of ocean climate change.

Global and regional circulation models of the ocean have the horizontal and vertical resolution required for realistic simulation of long-range acoustic propagation. These simulations offer a novel test of model accuracy, because acoustic propagation is sensitive to the vertical gradients of sound speed in the models and acoustic arrival patterns are known to have stable and universal properties. The travel time and dispersal of an arrival pattern are examples of such properties. Acoustic receptions on three long vertical line arrays from basin-scale transmissions in the North Pacific in 1996 and 1998 are used to test the acoustical properties of the time–mean state of several ocean models.

The NOAA World Ocean Atlas, a global representation of the "average" ocean, yields acoustic predictions whose patterns closely match the observations, but not all ocean models accurately represent oceanic sound speed properties. Acoustical tests of a model do not necessarily require actual data, because basic acoustical properties may be derived from the World Ocean Atlas.

During a 1960 experiment on long-range ocean acoustic propagation, the signals of explosive shots made near Perth, Australia were recorded by a hydrophone array at Bermuda. The acoustic paths followed by those pulses are recalculated using a modern atlas for global ocean sound speed. The blocking or refractive effects of topographic features, besides continents, are ignored in this simple calculation. No direct path to the hydrophone array is obtained, and the array is found to be in the acoustic shadow by 300–400 km. The large energy of the 300-lb explosive sound sources required for adequate SNR at the receiver also points to a shadowed reception. Given nominal estimates for temperature warming on the sound channel axis due to climate change, the experiment repeated today would give signals that arrive 12 s earlier, but this signal could not be reliably measured because of intra-annual variability.

Large-scale, range- and depth-averaged temperatures in the North Pacific Ocean were measured by long-range acoustic transmissions over the decade 1996–2006. Acoustic sources off central California and north of Kauai transmitted to receivers throughout the North Pacific. Even though acoustic travel times are spatially integrating, suppressing mesoscale variability and providing a precise measure of large-scale temperature, the travel times sometimes vary significantly on time scales of only a few weeks. The interannual variability is large, with no consistent warming or cooling trends.

Comparison of the measured travel times with travel times derived from (i) the World Ocean Atlas 2005 (WOA05), (ii) an upper ocean temperature estimate derived from satellite altimetry and in situ profiles, (iii) an analysis provided by the Estimating the Circulation and Climate of the Ocean (ECCO) project, and (iv) simulation results from a high-resolution configuration of the Parallel Ocean Program (POP) show similarities, but also reveal substantial differences. The differences suggest that the data can provide significant additional constraints for numerical ocean simulations. The acoustic data show that WOA05 is a much better estimate of the time–mean hydrography than either the ECCO or POP estimates and provide significantly better time resolution for large-scale ocean variability than can be derived from satellite altimetry and in situ profiles.

The barotropic Rossby wave field in the North Atlantic Ocean is studied in an eddy-resolving ocean model simulation. The meandering model Gulf Stream radiates barotropic Rossby waves southward through preferred corridors defined by topographic features. The smoother region between the Bermuda Rise and the mid-Atlantic Ridge is a particularly striking corridor of barotropic wave radiation in the 20–50 day period band. Barotropic Rossby waves are also preferentially excited at higher frequencies over the Bermuda Rise, suggesting resonant excitation of topographic Rossby normal modes. The prevalence of these radiated waves suggests that they may be an important energy sink for the equilibrium state of the Gulf Stream.

Large-scale temperatures in the North Pacific were measured by long-range acoustic transmissions from 1996–2006. Acoustic sources off California and Kauai transmitted to receivers distributed throughout the North Pacific from 1996–999. Kauai transmissions continued from 2002–2006. Acoustic travel time data are inherently integrating. This averaging suppresses mesoscale variability and provides an accurate measure of large-scale temperature, subject to the limitations of the ray path sampling. At basin scales, the ocean is highly variable, with significant changes occurring at time scales from weeks to years. The interannual variability is large compared to trends in the data. Willis, et al. used objective mapping techniques applied to satellite altimetry and hydrography to derive 0–750 m temperature fields for the global ocean. Travel times equivalent to the measured travel times can be calculated using these fields. The measured and calculated travel times are similar, but also show significant differences. Similar comparisions using travel times derived from the "Estimating the Circulation and Climate of the Ocean" (ECCO) model and a high-resolution Parallel Ocean Program (POP) model also show similarities and differences. The ECCO model was constrained by altimetric and profile data by data assimilation, suggesting that the acoustic travel times provide meaningful additional constraints on model behavior.

The past decade has seen a renaissance in the interest in oceanic internal tides and their role in mixing the abyssal ocean. One of the sparks of this renaissance was the discovery of the remarkable coherence of low-mode internal tides using an ocean acoustic tomography array deployed in the central North Pacific Ocean in 1987. The result was subsequently confirmed by satellite altimetry. This talk will review these results and related results from the 1991 Acoustic Mid-Ocean Dynamics Experiment (AMODE) in the North Atlantic and the 2001 Hawaiian Ocean Mixing Experiment (HOME) around the Hawaiian Ridge. Internal tides are generated by tidal forces hence they are a means by which the tides dissipate energy. With wavelengths of 150 km, low-mode semidiurnal internal tides appear to propagate 1000's of kilometers across ocean basins while retaining considerable coherence.

With several years of long-range (several Mm) acoustic propagation data obtained during the Acoustic Thermometry of Ocean Climate (ATOC) and North Pacific Acoustic Laboratory (NPAL) projects, the seasonal cycle of ocean temperature in the North Pacific can be examined. Acoustic transmissions have been made from a source located off the northern Californian coast and from a source located north of Kauai, HI to several receivers of opportunity located in the North Pacific Basin. The acoustic data are a high signal-to-noise measure of large-scale temperature with excellent temporal resolution. Although only a few realizations of the seasonal cycle are available, it is clear that inter- and intra-annual variabilities are large contributions to the time series, with signal amplitudes comparable to the seasonal cycle. Such variabilities are likely advective in origin. The time scales for some of the changes in temperature are short, sometimes of order weeks. Not all available acoustic paths are suitable for assessing the seasonal cycle, however. Near Hawaii, the acoustic sampling does not extend to the near-surface waters, so seasonal variations there are not measured. The acoustic results will be compared to measures of the seasonal cycle by satellite altimetry, profiling floats, and the ECCO numerical ocean model.

Mode-1 internal tides were observed the western North Atlantic using an ocean acoustic tomography array deployed in 1991%u20131992 centered on 25°N, 66°W. The pentagonal array, 700-km across, acted as an antenna for mode-1 internal-tides. Coherent internal-tide waves with O(1 m) displacements were observed traveling in several directions. Although the internal tides of the region were relatively quiescent, they were essentially phase locked over the 200–300 day data record lengths. Both semidiurnal and diurnal internal waves were detected, with wavenumbers consistent with those calculated from hydrographic data. The M₂ internal-tide energy flux was estimated to be about 70 W m^-1, suggesting that mode-1 waves radiate 0.2 GW of energy, with large uncertainty, from the Caribbean island chain at this frequency. A global tidal model (TPXO 5) suggested that 1–2 GW is lost from the M₂ barotropic tide over this region, but the precise value was uncertain because the complicated topography makes the calculation problematic. In any case, significant conversion of barotropic to baroclinic tidal energy does not occur in the western North Atlantic basin. It is apparent, however, that mode-1 internal tides have very weak decay and retain their coherence over great distances, so that ocean basins may be filled up with such waves. Observed diurnal amplitudes were an order of magnitude larger than expected. The amplitude and phase variations of the K₁ and O₁ constituents observed over the tomography array were consistent with the theoretical solutions for standing internal waves near their turning latitude. The energy densities of the resonant diurnal internal waves were roughly twice those of the barotropic tide at those frequencies.

The statistics of low-frequency, long-range acoustic transmissions in the North Pacific Ocean are presented. Broadband signals at center frequencies of 28, 75, and 84 Hz are analyzed at propagation ranges of 3252 to 5171 km, and transmissions were received on 700 and 1400 m long vertical receiver arrays with 35 m hydrophone spacing. In the analysis we focus on the energetic "finale" region of the broadband time front arrival pattern, where a multipath interference pattern exists. A Fourier analysis of 1 s regions in the finale provide narrowband data for examination as well. Two-dimensional (depth and time) phase unwrapping is employed to study separately the complex field phase and intensity. Because data sampling occured in 20 or 40 min intervals followed by long gaps, the acoustic fields are analyzed in terms of these 20 and 40 min and multiday observation times. An analysis of phase, intensity, and complex envelope variability as a function of depth and time is presented in terms of mean fields, variances, probability density functions (PDFs), covariance, spectra, and coherence. Observations are compared to a random multipath model of frequency and vertical wave number spectra for phase and log intensity, and the observations are compared to a broadband multipath model of scintillation index and coherence.

We examine statistical and directional properties of the ambient noise in the 10%u2013100 Hz frequency band from the NPAL array. Marginal probability densities are estimated as well as mean square levels, skewness and kurtoses in third octave bands. The kurotoses are markedly different from Gaussian except when only distant shipping is present. Extremal levels reached %u223C150 dB re 1 %u03BC Pa, suggesting levels 60dB greater than the mean ambient were common in the NPAL data sets. Generally, these were passing ships. We select four examples: i) quiescent noise, ii) nearby shipping, iii) whale vocalizations and iv) a micro earthquake for the vertical directional properties of the NPAL noise since they are representative of the phenomena encountered. We find there is modest broadband coherence for most of these cases in their occupancy band across the NPAL aperture. Narrowband coherence analysis from VLA to VLA was not successful due to ambiguities. Examples of localizing sources based upon this coherence are included. kw diagrams allow us to use data above the vertical aliasing frequency. Ducted propagation for both the quiescent and micro earthquake (T phase) are identified and the arrival angles of nearby shipping and whale vocalizations. MFP localizations were modestly successful for nearby sources, but long range ones could not be identified, most likely because of signal mismatch in the MFP replica.

In 1998–1999, a comprehensive low-frequency long-range sound propagation experiment was carried out by the North Pacific Acoustic Laboratory (NPAL). In this paper, the data recorded during the experiment by a billboard acoustic array were used to compute the horizontal refraction of the arriving acoustic signals using both ray- and mode-based approaches. The results obtained by these two approaches are consistent. The acoustic signals exhibited weak (if any) regular horizontal refraction throughut most of the experiment. However, it increased up to 0.4 deg (the sound rays were bent towards the south) at the beginning and the end of the experiment. These increases occurred during midspring to midsummer time and seemed to reflect seasonal trends in the horizontal gradients of the sound speed. The measured standard deviation of the horizontal refraction angles was about 0.37 deg, which is close to an estimate of this standard deviation calculated using 3D modal theory of low-frequency sound propagation through internal gravity waves.

Large-scale, depth-averaged temperatures have been measured by long-range acoustic transmissions in the North Pacific Ocean for the past nine years. Acoustic sources located off central California and north of Kauai transmitted to receivers distributed throughout the North Pacific from 1996 through 1999 during the Acoustic Thermometry of Ocean Climate (ATOC) project. The Kauai transmissions resumed in early 2002 and are now continuing as part of the North Pacific Acoustic Laboratory (NPAL) project; a six-year time series has been obtained so far. Even at long time and large spatial scales the ocean is highly variable. The paths from Kauai to California show a modest cooling trend (longer travel times) until the present time. A path to the northwest showed modest warming and a weak annual cycle from 1999 until early 2003, when a strong annual cycle returned. In retrospect, these changes stemmed from the warming of the central Pacific that occurred in this interval, possibly associated with the Pacific Decadal Oscillation (PDO).

Comparisons between measured travel times and those predicted using ocean models, constrained by satellite altimeter and other data, show significant similarities and differences. Comparison between upper-ocean Argo profiling float temperatures and the acoustically measured temperature along one path illustrates the strength of the integral measurements, with substantially lower uncertainty. The acoustic data ultimately need to be combined with sea-surface height Argo float data to determine the complementarity of the various data types. In particular, combining the acoustic and Argo data by inverse techniques will quantify the ability of the float data to resolve large-scale, upper-ocean heat content and the ability of the acoustic data to resolve abyssal temperature changes.

A method to obtain coherent acoustic wave fronts by measuring the space–time correlation function of ocean noise between two hydrophones is experimentally demonstrated. Though the sources of ocean noise are uncorrelated, the time-averaged noise correlation function exhibits deterministic waveguide arrival structure embedded in the time-domain Green's function. A theoretical approach is derived for both volume and surface noise sources. Shipping noise is also investigated and simulated results are presented in deep or shallow water configurations. The data of opportunity used to demonstrate the extraction of wave fronts from ocean noise were taken from the synchronized vertical receive arrays used in the frame of the North Pacific Laboratory (NPAL) during time intervals when no source was transmitting.

Acoustic transmissions from a 75-Hz source near Kauai to a vertical line array near California were recorded as part of the North Pacific Acoustic Laboratory (NPAL) experiment. Extensive environmental measurements were also performed as part of the experiment and were intended to ensure correspondence between numerical simulations and the data. Despite the availability of this information, the process of identifying the recorded arrivals with predictions has not been a simple one. Since the source is near the seafloor at about 800 m depth, and the depth at the receiver is approximately 1800 m, acoustic interaction with the bathymetry has been explored as a possible complication. Ray simulations that allow for specular reflection from the bottom indicate that fully-refracted and bottom-interacting paths can reach the receiver range (about 3900 km) at similar travel times. The simultaneous presence of both kinds of acoustic energy could contribute to the identification difficulties. A series of parabolic equation simulations have been performed for different geoacoustic parameters in an attempt to correspond more closely to the data. The sensitivity of the predictions to the method used to extract and interpolate the sound speeds has also been investigated.

In our previous paper [J. Acoust. Soc. Am. 112, 2232], we obtained a time dependence of the horizontal refraction angle (HRA) of acoustic signals propagating over a range of about 4000 km in the ocean. This dependence was computed by processing of acoustic signals recorded during the North Pacific Acoustic Laboratory (NPAL) experiment using a ray-type approach. In the present paper, we consider the results obtained in signal processing of the same data using a modal approach. In this approach, the acoustic field is represented as a sum of local acoustic modes with amplitudes depending on a frequency and arrival angle. We obtained a time dependence of HRA for a time interval of about a year. Time evolution of HRA exhibits long-period variations which could be associated with seasonal trends in the sound speed profiles. The results are consistent with those obtained by the ray approach. Different horizontal angles within arrivals were impossible to resolve due to sound scattering by internal waves. A theoretical estimate of the angular width of the acoustic signals in a horizontal plane was obtained. It appears to be consistent with the observed variance of HRA data.

Acoustic measurements of large-scale, depth-averaged temperatures are continuing in the North Pacific Ocean in a follow-on to the Acoustic Thermometry of Ocean Climate (ATOC) project. Long-range acoustic transmissions resumed in January 2002 from a low-frequency acoustic source located north of Kauai to U.S Navy receivers distributed throughout the North Pacific Ocean. The source previously transmitted for about two years (1997-1999) as part of the ATOC project. Both the source and receivers are connected to shore by cable, providing near-real time data. It is anticipated that transmissions will continue for five years, as part of the North Pacific Acoustic Laboratory (NPAL) project. At these ranges acoustic methods give integral measurements of large-scale ocean temperature that provide the spatial low-pass filtering needed to observe small, gyre-scale signals in the presence of much larger, mesoscale noise. The paths to the east, particularly those paths to the California coast, show cooling relative to the earlier data. A path to the northwest showed modest warming until early 2003, when rapid cooling occurred. The acoustic rays sample depths below the mixed layer near Hawaii, but extend to the surface near the California coast and north of the Subarctic Front. The acoustic data will be compared to and ultimately combined with upper-ocean data from ARGO and sea-surface height data from satellite altimeters to detect changes in abyssal ocean temperature and to test the complementarity of the various data types. Acoustic travel-time data have been used previously in simple assimilation experiments and are now shown in comparison with assimilation products from state-of-the-art efforts from the ECCO (Estimating the Circulation and Climate of the Ocean) Consortium.

The Acoustic Thermometry of Ocean Climate (ATOC) provided an opportunity to observe signals propagating in the low-order modes of the ocean waveguide. Understanding the fluctuations of these mode signals is an important prerequisite to using them for tomography or other applications. In previous work, we characterized the cross-mode coherence and temporal variability of the low-order mode arrivals at 3515 km range [Wage et al., J. Acoust. Soc. Am. (in press)]. This study compares the mode arrivals for two different ranges: 3515 km and 5171 km, using data from the ATOC vertical line arrays at Hawaii and Kiritimati. We discuss the mode intensity and coherence statistics for each of the arrays and examine mean arrival time trends over the year-long deployment. Experimental results are compared to PE simulations of propagation through a realistic background environment perturbed by internal waves of varying strengths. The dependence of mode statistics on the path-dependent changes in the background sound speed and the parameters of the internal wave field is explored.

A MATLAB-based MAP graphical user interface (GUI) is described for accessing bathymetry and sound speed databases and using them to predict ocean acoustic transmission and to display the results. MATLAB scripts have been developed to estimate a field of view from a single transducer and to compute transmission losses between a source and receiver using one of several standard propagation codes. The GUI called MAP is described and several test cases are presented. The MAP GUI supports three simple functions that display database information; an image of bathymetry for a user-specified area may be plotted, a sound speed profile (SSP) at a user-selected location may be plotted, and bathymetry and SSP along a geodesic between a transmitter and receiver may be plotted. The MAP GUI supports two types of algorithms modeling acoustic transmission loss. The first is computation of the acoustic field between two points. External propagation codes written by other agencies are used to do the basic simulation. MATLAB accesses database information, writes input files for the simulations, runs them, and plots results. Currently five external simulations are supported; (1) EIGENRAY, (2) RAY, (3) CASS, (4) UMPE, and (5) RAM. The first three simulations are ray-based models, and the last two are parabolic expansion models. The second type of algorithm is mapping a "field of view" for a transducer (a "shadow" plot). Horizontal rays are propagated from the instruments and the range at which acoustic propagation fails is displayed. Two algorithms are supported. One accesses a database of sound channel depths along with bathymetry information and uses heuristics to estimate propagation along the horizontal rays. The other accesses the Levitis SSP database and uses Weston's Ray Invariant to identify ray elevations "interrupted" by bathymetry or (if specified by the user) by the surface.

As part of the North Pacific Acoustic Laboratory project, ambient sound data from 1994 to the present has been collected. Long-term averages of these data from a receiver on the continental slope west of Point Sur, CA, are compared to earlier measurements made at the same site over 1963–1965 by Wenz [Wenz, J. Underwater Acoust. 19 (1969)]. The levels Wenz reported fall below our 10% quantile from 5 Hz to 50 Hz, rise to the 50% quantile (i.e., the median) at 100 Hz, and again fall below the 10% quantile by 250 Hz. Wenz removed highly variable "transient" data before calculating his averages. We mimicked his processing with the NPAL data and obtained a result which is virtually indistinguishable from the median, which is approximately 1 dB below the (dB) mean of each one-third octave band. Hence, our median levels are directly comparable to Wenz's results, and this comparison shows that the 1994–2000 levels exceed the 1963–1965 levels by 9 dB or less below 100 Hz and again at 250 Hz, but are roughly similar at 100 Hz.

The low-order modes constitute some of the most energetic arrivals at long ranges. Understanding fluctuations of these mode arrivals is crucial to their use as observables in matched field processing and tomography. Both simulated and experimental data indicate that at megameter ranges, the low modes have complex arrival patterns due to internal-wave-induced coupling. Analysis of broadband receptions at 3515 km from the ATOC experiment has shown that mode coherence times are on the order of 6 minutes and that centroid statistics provide useful measures of arrival time trends over the course of several months [Wage et al., IEEE Sensor Array and Multichannel Signal Processing Workshop Proceedings, pp. 102-106, 2000]. The North Pacific Acoustic Laboratory (NPAL) experiment presents an opportunity for further research on broadband mode arrivals at megameter ranges. This study examines temporal coherence, intensity variations, and other mode statistics using data from the 40-element NPAL vertical line array. Experimental results are compared with PE simulations of propagation through internal waves of varying strengths, and the impact of the up-slope propagation near the receivers on the mode statistics is discussed.

Low-frequency (75-Hz) acoustic signals were repeatedly transmitted over a 1 year period and sampled vertically (with up to a 1400-m aperture) and horizontally (with a 3600-m cross-range aperture) by a distant billboard array (3900-km range) as described by the NPAL Group. The data are complicated by the fact that the sound interacts with the bottom near both the source and receiver. Vertical beamforming is used to filter the bottom interacting energy, and thus allow analysis of the fundamental acoustic properties. Subband and subarray processing is used to produce estimates of arrival times and angles or resolved ray arrivals. Time-series of acoustic travel times and arrival angles are then developed.

A comprehensive, long-range sound propagation experiment was carried out with the use of the billboard acoustic array of the North Pacific Acoustic Laboratory (NPAL) in 1999. The antenna consisting of five vertical line arrays was deployed near a California coast and received broadband acoustic signals transmitted from Hawaii over a distance of about 4000 km. Acoustic signals propagating over such a long distance might exhibit noticeable horizontal refraction. The paper will present results of processing the NPAL data to infer horizontal refraction angle (HRA) as a function of time. Different methods were used for determining HRA. The first approach employed cross correlation of the acoustic signals at different VLAs. Time delay corresponding to maximum of cross correlation is related to HRA assuming the angle is approximately the same for all rays (or modes). The second method used modal representation of the arriving broadband signals. The dependency of the amplitudes of acoustic modes on mode number, frequency, and arrival angle was determined independently within narrow frequency bins, and then the results were averaged over whole frequency range. This method allowed, in particular, to evaluate angular width of the arrived signal, which appeared to be of the order of a few milliradians.

While the NPAL array was primarily deployed to examine the spatial coherence of the Hawaii source, it is also a rich data set for ambient noise studies. Shipping noise, earthquakes and biologics all have been identified in the NPAL data. Moreover, ambient noise coherence is the primary issue in maximizing the SNR output of a sonar system. The first and second order statistics of data from the NPAL "noise only" segments have been analyzed with the following results: (i) There is a wide spread in the peak levels, most likely due to the proximity to shipping lanes. The maximum peak level in the recording band is 117 dB. (ii) Full broadband coherences tend to be low because of the presence of many ships. (iii) If one examines frequency bands of 1–2 Hz, then lines of individual ships can be identified and associated and they are very coherent across NPAL aperture. (iv) Vertical beamforming indicates relatively highly directional spectra at low grazing angles and "noise notch" for the spectra at higher frequencies. Horizontal beamforming has been difficult to implement due to element positioning errors and the large array transit time.

One of the main objectives of the NPAL experiment is to investigate the horizontal refraction and coherence of the acoustic wave fronts at long range. Given time series of acoustic arrival times and angles of resolved ray arrival arrivals, a detailed look at the acoustic wave fronts is possible. First and second order statistics (density functions and coherences) of the wave fronts are investigated. The wave fronts are shown to vary with time, frequency, depth and across the horizontal aperture.

Receptions of long-range acoustic transmissions by deep hydrophone arrays in the Pacific and Atlantic often have "ray-like" arrivals that occur in the shadow zone of the predicted time front. These "ray-like" arrivals can frequently be identified with the cusps of the predicted time front, but the receivers are up to 750 m below the depth of the cusps. Preliminary calculations show that the observed acoustic energy is not accounted for by errors in the sound speed, leakage of acoustic energy below the cusps as predicted by the full wave equation, or scattering due to internal waves. Data obtained during experiments in the Atlantic and Pacific will be reviewed. Experiments that have been conducted with receivers of vertical line arrays have not had receivers deep enough to observe this phenomena. The effect is seen when bottom-mounted or midwater acoustic sources are used. These data present a number of problems: If the ray paths are wandering all over the water column, why are predictions of ray travel times usually accurate? How does the energy loss associated with these data increase the attenuation of very long-range acoustic transmissions? Without knowing the forward problem, how can these data be used to determine oceanographic information?

The North Pacific Acoustic Laboratory program augmented the existing ATOC acoustic network with a sparse, two-dimensional receiving array installed west of Sur Ridge, California, from July 1998 through June 1999, to receive transmissions from the 75-Hz ATOC source north of Kauai. The NPAL array consisted of four 20-element vertical arrays, each with a 700-m aperture, and one 40-element vertical array with a 1400-m aperture. The arrays were deployed transverse to the 3900-km acoustic path from the Kauai source and had a total horizontal aperture of 3600 m. Data collected with the billboard array and the U.S. Navy SOSUS receivers are being used (i) to study the temporal, vertical, and horizontal coherence of long-range, low-frequency resolved rays and modes, (ii) to study 3-D propagation effects, (iii) to examine directional ambient noise properties, and (iv) to improve basin-scale ocean nowcasts via assimilation of acoustic data and other data types into models. In addition to acoustic data, environmental data along the path from the Kauai source to the billboard array were acquired by two oceanographic sub-surface moorings and by two XBT/CTD/ADCP transects along the path. The experiment will be described and some preliminary results presented.

Acoustic transmissions on basin scale ranges are being used to determine depth-dependent temperature variability. With travel time being the primary observable, stationary sources and nearly stationary receivers are experimental requirements. This has led to the use of bottom-mounted sources and receivers to reduce travel time variability. The NPAL (North Pacific Acoustics Laboratory) experiment has transmitted broadband acoustic pulses from two bottom-mounted sources near the sound channel axis. Recordings have been taken on the NPAL billboard array, a linear series of five vertical line arrays moored in 1800 m of water near Monterey, CA. Additional recordings have been taken from the SOSUS system throughout the Pacific basin. The effects of the near source and near receiver scattering are examined. In particular, near source scattering leads to excess high-angle energy entering deep water with a travel time delay of nearly 1 s due to the low group speeds of high-angle rays/modes in shallow water. We also compare the energetics of the arriving rays that have bounced once on the rising seafloor near the NPAL receivers. Comparisons of models and data for bottom interacting acoustics lead us to the perennial issue of geoacoustic parameters.

Time series of temperature have been measured acoustically in the northeast Pacific as part of the Acoustic Thermometry of Ocean Climate (ATOC) project. These time series are compared with other available data types. The acoustic time series of transmissions from the California and Kauai acoustic sources were obtained during 1996–1999. As a result of marine mammal protocols, the time series are intermittent; the California source was turned off in Fall 1998. Assuming that variations in sea-surface height observed by TOPEX/POSEIDON are caused by thermal expansion, the amplitude of the annual cycle of heat content derived from altimetry is larger than that found by the acoustic data, Levitus climatology, and monthly maps of ocean temperature from XBTs of opportunity. The heat content "anomalies" determined by the XBT maps are comparable in size to the differences between the XBT and acoustically derived heat content. A variety of problems with the XBT sampling may account for these differences. The 12-year time series of temperature derived from the Hawaiian Ocean Time series (HOT) data highlights the mesoscale noise in single-point sampling. However, thermal variability at 100-day time scales is observed in the acoustic data obtained between Hawaii and California using the Kauai source. Acoustic thermometry is complementary to altimetry and hydrography.

The North Pacific Acoustic Laboratory program augmented the existing ATOC acoustic network with a sparse, two-dimensional receiving array installed west of Sur Ridge, CA, close to an existing U.S. Navy SOSUS array, during July 1998 to receive transmissions from the 75-Hz ATOC source north of Kauai. The NPAL array consisted of four 20-element vertical arrays, each with a 700-m aperture, and one 40-element vertical array with a 1400-m aperture. The arrays were deployed transverse to the 3900-km path from the Kauai source and had a total horizontal aperture of 3600 m. Data collected with the two-dimensional array and the U.S. Navy SOSUS receivers will be used to (i) study the temporal, vertical, and horizontal coherence of long-range, low-frequency resolved rays and modes, (ii) study 3D propagation effects, (iii) examine directional ambient noise properties, and (iv) to improve basin-scale ocean nowcasts via assimilation of acoustic data and other data types into models. Environmental data along the path from the Kauai source to the two-dimensional array were acquired by two oceanographic subsurface moorings and by two XBT/CTD/ADCP transects along the path, one at the beginning and one at the end of the experiment. We describe the experiment and offer some preliminary data.