The University of Alaska Fairbanks (UAF), with Science Applications International Corporation (SAIC), the Massachusetts Institute of Technology (MIT), the NOAA Environmental Technology Laboratory (ETL), the Shirshov Institute of Oceanology, Moscow, and the Institute of Applied Physics (IAP), Nizhny-Novgorod propose to:

• recover and replace an autonomous acoustic/oceanographic mooring that has been operating in the Lincoln Sea since October 1998 (research themes 1, 5,and 6),
• analyze the 18+ months of average basin scale Arctic Ocean temperature between Franz Victoria Strait and the Lincoln Sea derived from measured modal acoustic travel times and in-situ CTD data from this array, a unique and unprecedented data set of extraordinary value for detection of contemporary climate change (research themes 1, 5, and 6),
• compare and assimilate these data with/into coupled atmosphere/ice/ocean models being run at UAF (research themes 1, 3, 5, and 6),
• conduct research in tomographic inversion of modal travel times for integrated sound speed as a function of depth, and acoustic remote sensing of ice thickness and roughness, and the thickness of the Arctic Ocean mixed layer (research theme 1), and
• perform a design study to develop the next generation source mooring for acoustic remote sensing in the Arctic Ocean (research theme 1).

UAF with SAIC will provide overall project coordination and management. UAF will lead the atmosphere/ice/ocean modeling, data assimilation and climate analyses. SAIC will lead the array recovery operation in the fall of 2000 in the Lincoln Sea, will build the replacement autonomous array, and will deploy the replacement array. ETL, SAIC, Shirshov, and MIT will perform the oceanographic and acoustic remote sensing analysis, simulation, and modeling research. The Institute of Applied Physics will perform the source array design study. The ATAM will advance our capabilities for expanded observational capabilities throughout the Arctic Ocean, exploiting acoustic remote sensing and long-term autonomous and cabled undersea moorings at a time when unprecedented Arctic change has been documented. The need for hard data on global climate change is a national issue. Despite the pressing need to improve the observational capabilities in the Arctic, there are substantial logistical costs inherent in the proposed research. We remind our reviewers that arctic logistics add to the total cost of this proposal. Such costs, however, are necessary when improving the science of arctic observations.

It is known today that the Arctic ice/ocean/atmosphere system is undergoing dramatic changes and is intrinsically a more dynamic and variable system than previously understood. Since the early 1990s, the average temperature in the core of the Atlantic Layer has increased by approximately .9° C with major frontal features showing as much as 1.4° C increases (Fig. 1). Analysis of satellite passive microwave images shows a 3% per decade decrease in sea ice extent since 1978, with a more rapid decline of 4.3% between 1987 and 1994 [Bjorgo et al. 1997, Johannessen et al. 1995, 1996]. Finally, a large decrease of the annual mean atmospheric sea level pressure over much of the Arctic has been observed in this decade. In April 1994, acoustic transmissions were made from a site north of the Svalbard Archipelago across the entire Arctic Ocean to receiver arrays located in the Lincoln Sea and the Beaufort Sea as part of the Transarctic Acoustic Propagation (TAP) Experiment. TAP was the first basin-scale observation of warming of the Atlantic Layer of the Arctic Ocean [Mikhalevsky et al. 1995a, 1995b, 1996, and 1999] (Fig. 2). The 1993 cruise of the USS Pargo [Morison et al. 1996ab] and the cruise of the CCGS Henry Larson [Carmack et al. 1995a; McLaughlin et al. 1996] revealed a shift in the boundary between the warmer more saline Atlantic water and the Pacific water towards the Alpha and Mendeleyev Ridges, and warmer water in the Makarov Basin, respectively. The Arctic Ocean Section of the USCGS Polar Sea and the CCGS Louis S. St Laurent [Carmack et al. 1996] conducted in August 1994 confirmed these results. Whether these recently observed changes in the Arctic ocean/atmosphere/ice system are a manifestation of a secular global climate change trend with an anthropogenic fingerprint [Overpeck et al. 1997, Vinnikov et al. 1999] or a "natural" oscillation [Grotefendt et al. 1998, Johnson et al. 1999] is an area of active research. Recent modeling has suggested that major shifts in the Arctic Ocean circulation occur on a decadel time scale between two dominant circulation regimes. A key element of decadel variability appears to be the quantity and timing of Atlantic Water input to the Arctic Basin via Fram Strait and the Barents Sea. These variations appear to be associated with the wind-driven component of the Arctic Ocean circulation which has been shown to alternate between anticyclonic and cyclonic [Proshutinsky and Johnson 1997; Johnson et al. 1999, Polyakov et al. 1999]. The acoustic thermometry measurements from the Lincoln Sea will be used to estimate oceanic heat content in the Arctic Ocean north of the Fram Strait from which we can infer a suite of climatically relevant characteristics including air temperature, precipitation, ice motion and thickness, and permafrost temperatures over the surrounding land masses.

Fig. 1 SCICEX 1998 submarine CTD sections upper panel along a trans-Arctic path from the Chukchi Sea to the Eastern Arctic with climatology from the U.S./Russian Joint ATLAS [EWG, 1997] middle panel subtracted out showing the temperature difference, lower panel.

Fig. 2. Geographic positions for the TAP experiment, April 1994, Submarine SCICEX tracks and ACOUS source and receiver positions. The Lincoln Sea array is autonomous. The Barrow Receive Array is the vertical line array mooring of the BAO located approximately 150 kms off Barrow.

The importance of the Arctic in the Earth's global climate system is well understood. Yet, despite the Arctic's role as a sensitive indicator of global climate change, and the mounting evidence of major changes that are occurring now, its internal processes remain poorly understood. Our sparse knowledge about the water mass structure of the Arctic is based mainly on sporadic measurements made at drifting ice-camps and oceanographic surveys which used aircraft landings onto the ice [Gorshkov 1980; Treshnikov 1977, 1985]. An effective oceanographic survey section scheme of ocean dynamics investigations had never been done in the Central Arctic before 1987 because the Arctic Basin is one of a few areas in the world's oceans which are virtually inaccessible for ordinary ocean-going research vessels. Until now, only eleven sections have been obtained in the Arctic Basin with the help of icebreakers (Polarstern – 1987, Rossiya – 1990, Oden – 1991, Henry Larsen and Polar Star – 1993, Louis S. St. Laurent and Polar Sea – 1994) and submarines (Pargo – SCICEX-93, Cavalla – SCICEX-95, Pogy – SCICEX-96, Archerfish – SCICEX-97, and Hawkbill – SCICEX-98 and 99). Only six of these have been long trans-basin sections, the Louis S. St Laurent and Polar Sea in 1994, and the SCICEX 1995-1999 cruises.

Understanding the complex feedback mechanisms and external forcing at work in the Arctic that drive its natural variability and determine its response to greenhouse warming is hampered, in large part, by a lack of synoptic field measurements with incomplete and undersampled records in time and space. The Arctic is not a major focus of the World Ocean Circulation Experiment (WOCE) program, nor was it included in the report, "Toward a U.S. Plan for an Integrated, Sustained Ocean Observing System," prepared for and submitted to Congress on April 26, 1999, by the National Ocean Research Leadership Council (NORLC) of the National Oceanographic Partnership Program (NOPP). The only major coordinated Arctic research experimental program, other than SCICEX and ACOUS, is the SHEBA project which completed its field program in August 1998. The SCICEX 1999 cruise is the last dedicated use of a submarine for Arctic research planned for the foreseeable future. While satellite remote sensing is important for measuring the extent and providing information on sea ice type in the Arctic, satellites cannot provide any visibility into the Arctic Ocean itself. The need to exploit new unmanned technologies to create sustained long-term observations in the Arctic Ocean and integrate these observations with satellite data is required. ATAM will begin to fill this observational gap by providing the first year-round basin scale measurements using acoustic thermometry and in-situ measurements from the Lincoln Sea autonomous array that will be recovered in the Fall of 2000.

ATAM will realize considerable leverage of both the scientific expertise and experience from the ACOUS project and the research team that has been working together since the early 1990s. Research on acoustic thermometry in the Arctic was begun in 1990 by SAIC as an outgrowth of the Acoustic Thermometry of Ocean Climate (ATOC) Program [Mikhalevsky et al. 1991]. MIT and WHOI were early and on-going partners. The project was formalized under the name ACOUS as a collaborative program with Russia in 1994 under the auspices of the Gore-Chernomyrdin Commission (now Gore-Stepashin Commission) and a Memorandum of Understanding was signed by the U.S. Secretary of Defense, the Russian Deputy Minister of Defense, and the Russian Minister of Science. UAF, the Barrow Arctic Science Consortium (BASC), and the Department of Wildlife Management, North Slope Borough joined the ACOUS effort in 1996. The Russian participation has and will continue to be important to the success of ACOUS. Their participation has permitted the ACOUS program access to Russian Arctic territory for experiments, to Russian research institutes and researchers and the vast store of Russian data, experience and technical expertise in the Arctic, previously unavailable to western researchers. The Institute of Applied Physics (IAP) in Nizhny Novgorod has been responsible for the Arctic acoustic source design. They successfully designed, built and deployed the 20 Hz source used during the TAP experiment in April 1994. This experience went into the design of the autonomous source that is currently operating in the Arctic. Data acquired at an ice camp established in April 1999 have shown that the source is operating exactly according to specifications. The U.S. Navy has independently confirmed the source operation. The Shirshov Institute of Oceanology, and the Andreev Acoustics Institute in Moscow have been key performers on the program and the lead Russian Principal Investigator, Dr. Alexander Gavrilov is from Shirshov. Important analyses and access to Arctic Ocean data bases has been provided by the Arctic Antarctic Research Institute (AARI) in St. Petersburg and they also provided ice breaker services of the Akademik Fedorov in September and October 1998.

Under the provisions of the US/Russia 1994 MOU, an Intergovernmental Working Group and a Science Subgroup were formed and they have met regularly over the past five years. The U.S. Co-Chairman of the Intergovernmental Working Group is Dr. Ralph Alewine III (DASN for Nuclear Treaty Programs) and the Russian Co-Chairman is Rear Admiral Grigory Korl'kov (Russian Navy). The U.S. Co-Chairman of the Science Subgroup is Dr. Peter Mikhalevsky (SAIC), and the Russian Co-Chairman is Prof. Sergei Lappo (Director, Shirshov Institute of Oceanology, Moscow). The commitment of the Russian side to their obligations under the MOU and Working Group agreements has been unwavering. At the height of the financial crisis in Russia in August 1998, the Ministry of Science and the Russian Academy of Sciences intervened to ensure that the funds necessary for the Russian icebreaker Akademik Federov were available, so it could get underway and deploy the ACOUS source per the agreed ACOUS Science Plan (Fig. 3).

Fig. 3. ACOUS source being installed from the Russian icebreaker, Akademik Fedorov, in the Franz Victoria Strait, Oct. 9, 1998.

ATAM will benefit from the investment that has already been made in the ACOUS Project. This includes of course the data currently being collected on the acoustic/oceanographic array that was deployed in the Lincoln Sea coincident with the source deployment in October 1998. The U.S./Russian investment in the source and receive arrays including the Russian contribution for the source deployment by the Akademik Fedorov amounts to a total of $3 million. The continued collaboration with Russia is expected to yield additional opportunities for shared costs, including Russian installation of the replacement source in the Franz Victoria Strait in 2001. This year the Russian government provided $150,000 of funding to support researchers at the Shirshov Institute and IAP to support ACOUS research from their program "Research and Development for Civil Purposes in Priority Areas of Science and Technology."

The ACOUS project has also been working with the European Commission funded program called Acoustic Monitoring of the Ocean Climate in the Arctic. This project was started in 1997, spurred by the success of the TAP experiment, and is currently in a research phase with a primary focus on the use of acoustic remote sensing to determine ice thickness and correlate the acoustic data with satellite remote sensing data. The project leader is Prof. Ola M. Johannessen, Director, Nansen Environmental and Remote Sensing Center (NERSC) and includes the University of Cambridge, Scott Polar Research Institute (UCAM-SPRI), the Max Planck Institute for Meteorology (MPI), and the Nansen International Environmental and Remote Sensing Center (NIERSC), Russia. Researchers from UCAM-SPRI were aboard the Polar Star in June/July 1999, and recorded signals from the ACOUS source. Synergy is expected with this project when it enters its field program phase in the coming years.

Performers: UAF, ETL, SAIC, MIT, Shirshov

This task will include the recovery of the Lincoln Sea Array in the fall of 2000, and the processing, analysis, and dissemination of the acoustic and physical oceanography data. The acoustic signal processing will include beamforming, modal detection, and pulse compression processing and phase detection to determine the arrival times of the pulses as a function of acoustic mode. These steps are straightforward and have already been successfully demonstrated in the Arctic during the Transarctic Acoustic Propagation (TAP) experiment in April 1994 [Mikhalevsky et al. 1995a; Gavrilov and Mikhalevsky 1996; Mikhalevsky et al. 1999], and from data just recently collected at the APLIS Ice Camp in April 1999 in the Chukchi Sea. The travel-time data (one 20.7-min sample every four days) will be converted to temperature and heat content change and compared to model runs (see (4) below) and SCICEX CTD transects. These data combined with the SCICEX transects, the TAP measurement and the current data being taken in the Lincoln Sea will provide a total time series of over a decade. The theory that the Arctic is cycling or oscillating between two regimes based upon model runs [Proshutinsky and Johnson 1997, Johnson et al. 1999] and historical data analysis [Grotenfendt et al. 1998] suggest a five to seven year cycle, and there is evidence that the cyclonic regime of the early 1990s is or has already shifted to its anticyclonic state starting in 1998; however, it is not clear whether this has manifested itself in changes in the cooling of the Atlantic Layer of the Arctic Ocean. In fact, recent data from the acoustic section taken at the APLIS Ice Camp in April 1999 indicate that temperatures have remained the same or may have even increased slightly since 1998, but more careful analysis is needed to confirm this. The data from the Lincoln Sea array will provide the observations needed to understand the phasing of atmospheric and ocean changes in the Arctic and test the Arctic Oscillation theories, specifically the regime shift, predicted to have started in 1998.

The oceanographic structure of the Arctic Ocean makes it especially suitable for acoustic monitoring. The half-channel acoustic duct with the upward refracting sound speed profile creates a propagation condition in which the acoustic modes extend from the surface deeper into the water column as mode number increases. At the operating frequency of 20 Hz mode 1 is trapped above the thermocline and propagates in the upper mixed layer above the Atlantic Intermediate Layer. Mode 2 extends into and propagates almost entirely within the Atlantic Intermediate Layer, while modes 3 and higher sample the deeper depths in the Arctic Ocean. This has been shown with modeling [Gavrilov and Mikhalevsky 1995] and from analysis of the TAP data [Mikhalevsky et al. 1995a; Mikhalevsky et al. 1999].

This unique aspect of acoustic propagation in the Arctic also causes temporal dispersion of the modes. Mode 1 travels in the slowest sound speeds associated with the upper mixed layer and successive modes extending deeper are propagating at faster group velocities due to the increasing sound speed with depth. The 2,600 km propagation path from Turpan, the TAP source camp north of Svalbard to SIMI, the receive array camp in the Beaufort Sea (Fig. 2), resulted in a temporal separation of approximately 15 secs between mode 1 and mode 2, and approximately 2 secs between mode 2 and mode 3. This permits tracking of the travel-time changes for each mode, and observations of the temperature changes that can be associated with the important vertical stratifications within the Arctic Ocean. Indeed, it was the measurement of a 2 sec faster travel time of mode 2 from the TAP experiment when compared to a modeled result using historical climatology that calculated to an average increase in temperature of the Atlantic Intermediate Layer along the TAP propagation path of 0.4° C. Subsequent direct measurements from Arctic sections taken by the icebreaker CCGS Louis S. St-Laurent and the transect taken by the USS Cavalla (SSN-684) confirmed this result [Mikhalevsky, Keenan and Baggeroer 1996]. The largest temperature change, up to 1ēC, was observed in the core of the Atlantic water, over the Lomonosov Ridge, which may be due to change in the water inflow from the North Atlantic. A considerable decrease of the heat loss in the Norwegian Sea is assumed to be one of the causes of that phenomenon [Carmack et al. 1995b]. The TAP measurement was made in April 1994. As discussed above, another measurement was made in April 1999 from the APLIS Ice Camp with the ACOUS source installed in October 1998 and this travel time corrected for the range difference between the two experiments showed a decrease in travel time of an additional 2.6 secs since the April 1994 measurement or an increase of the average temperature in the core of the Atlantic Layer of another .5° C for a total of .9° C since April 1994. This result is consistent with the SCICEX 1998 section shown in Fig. 1 and actually indicates that there may be an additional .1° –.2° C increase since 1998. These data are being analyzed now so results are preliminary and will be compared to the SCICEX 1999 CTD section (SAIC received the data in July 1999) which was taken along the acoustic propagation path along the same track as SCICEX 1998 (see Fig. 2).

Data from the propagation path from Turpan to the Lincoln Sea during the TAP experiment has also been extensively analyzed. Due to the shoaling of the path into the Lincoln Sea to depths substantially less than the turning depths of modes 2–5, modal coupling occurs. This can result in acoustic energy traveling at different group velocities associated with the amount of time the energy "spends" in each mode. Travel-time changes associated with this phenomenon must be accounted. Recent analysis has shown that using the modal phase to measure the travel-time changes [Gavrilov and Mikhalevsky 1996] reduces the sensitivity to changes due to modal coupling [Gavrilov and Mikhalevsky 1999]. Travel time errors for the Lincoln Sea path due to these effects should not exceed an equivalent temperature measurement error of 5m° C [Gavrilov and Mikhalevsky 1999]. While the Beaufort Sea array will be placed in deep water to avoid any mode coupling problems, future source and receive sites could well be located on some of the shallower shelf regions in the Arctic. The techniques developed here can be used to assess various sites to insure that mode-coupling effects can be properly accounted for.

Another unique aspect of the acoustic channel in the Arctic is its remarkable phase stability. This was first observed in 1981 over a 330km path in the Eastern Arctic [Mikhalevsky, 1981]. This was again observed in 1994 during the TAP experiment over the 2,600km path from Turpan to SIMI [Mikhalevsky et al. 1995a; Mikhalevsky et al. 1999] and to the Lincoln Sea [Pawlowicz et al. 1996; Gavrilov and Mikhalevsky 1999]. This stability is a consequence of the order of magnitude less energy in the Arctic internal wave, and mesoscale fields compared to the temperate oceans. Over the 2,600 km propagation path in TAP, these perturbations were less than .5msec rms, or equivalently .1m° C. Thus travel time measurements using the phase of standard tomographic Maximal Length Sequences at 20 Hz which were transmitted during the TAP experiment and which are currently being used in ACOUS have accuracies less than 1 msec in travel time (~.1m° C) for the open ocean paths, and 5–10 msecs (~5m° C) over shoaling paths like the path to the Lincoln Sea. The temperature differences shown in Figure 1 and the resulting trends are easily detected (see a complete analysis in Mikhalevsky et al. 1999) and would have been detected had an acoustic thermometry grid been in place in the early 1990s when these changes began. Modeling runs for the travel time differences along the long transarctic path associated with the Arctic Ocean circulation regime shifts proposed by Proshutinsky and Johnson [1997] are approximately 1 sec., and seasonal variations are estimated to be approximately .1 secs. All of these signals are well above the measured travel-time noise level of approximately 5–10 msec and will be easily detected and tracked by the acoustic measure.

Performers: SAIC, MIT, ETL, Shirshov

The goal of this research is to develop a reliable measure of average ice thickness along acoustic paths in the Arctic Ocean to provide a measure of sea ice volume, a critical input for global climate models for which no affordable technique exists today.

The first objective is to quantify the acoustic attenuation as a function of the important ice parameters, including, roughness, thickness, and correlation lengths. This will involve using principally three ice loss models as discussed below to solve this forward problem. The goal is to quantify the achievable resolution of thickness and roughness changes that can be derived from the measured changes in attenuation. The use of submarine ice profile data and satellite remote sensing data to provide ground truth with the measured ACOUS 20 Hz data will be important for this research.

The second objective is to examine the use of broadband or multi-frequency approaches and other systems concepts that might use drifting transceiver buoys for example with separations of hundreds, vice thousands of kms to permit the use of higher frequencies that will be more sensitive to the ice parameters. At 20 Hz, the ice cover has no effect on the phase or travel time of the acoustic signals, only on the attenuation. At higher frequencies, greater than 100 Hz, not only is the attenuation much greater over the same range, but changes in travel time could be another observable.

Although the intensity of an acoustic signal is diminished by the interaction with the ice surface, the detailed physics of the interaction is still an open research question. To use the ACOUS signal to monitor the average ice roughness and ice thickness, we consider three ice loss models. The simplest ice loss model is the U.S. Navy standard ice loss model [Keenan 1989] that is empirical in nature and is only sensitive to the standard deviation roughness of the under-ice draft averaged in 1 nm segments. The Russian [Gavrilov et al. 1995] and LePage-Schmidt [Lepage and Schmidt 1994] ice loss models are both physics based models but each makes different assumptions about the properties of the ice surface and each require a different set of ice statistics. Some coincident acoustic measurements with measured ice draft data are available. When the parameters required for each of the models are computed from the measured ice draft data, the models show very good agreement among themselves and with data. The physics based ice loss models relate the ice loss, hence acoustic intensity degradation, to both the ice roughness and ice thickness that we wish to monitor acoustically. The Russian ice loss model incorporates a two-scale model of the ice structure associated with the ridges and flat ice that characterize real sea ice morphology, and may provide the best approach for attenuation modeling, given good ground truth data. Indeed, the significant limitation in making more progress on relating the physical parameters of the ice to modeled results has been the lack of a suitable data base of coincident acoustic propagation and ice draft, and ice type measurements. Digital Ice Profiling Sonar (DIPS) data taken during the April 1999 submarine SCICEX cruise and the acoustic transmissions recorded at the APLIS Ice Camp will provide such coincident data. The use of satellite remote sensing data including RADARSAT will be employed to assist in providing some ground truth for ice type and extent as well.

Performers: SAIC, MIT, WHOI, Shirshov

This research involves analyzing the capability of using a multiple frequency acoustic remote sensing technique to reliably determine the depth of the Arctic thermocline. The analysis will also determine the resolution that can be obtained using the 20 Hz ACOUS signal through observation of changes in the group velocity of mode 1 from the Lincoln Sea data. Because much of the interannual variability in the Arctic Ocean temperature is confined to the upper layers, it will be important to quantify effects associated with thermocline changes and other variability.

The accurate mode arrival time measurements that will be made by the ACOUS experiment vertical line array can potentially be used to determine the depth of the main thermocline in the Arctic Ocean. The thermocline, a small region characterized by steep temperature gradients, separates the warm Atlantic intermediate water and the cold Arctic subsurface water and mixed layer. There have been some indications from data taken during the 1995 and 1996 SCICEX cruises in the Arctic that the upper mixed layer and the associated thermocline may be weakening [Steele and Boyd 1998]. The thickness of the upper mixed layer is important because the surface waters act as a thermal buffer between the warmer Atlantic intermediate water and the Arctic ice cover. Changes in this upper layer could portend major shifts in the Arctic Ocean, including an ice-free state.

The following method will be explored to measure the average depth of the thermocline. For low frequency (20 Hz) sound, the acoustic normal mode 1 is trapped in the surface duct. Spending its time in the cold and slow surface waters and interacting with the ice canopy will cause it to arrive significantly later and more attenuated then the other modes that are not trapped. As the frequency of the acoustic signal goes up, mode 2 and then mode 3 are eventually trapped. By measuring the frequency at which modes 1, 2 and/or 3 are trapped in the surface duct, the depth of the thermocline can be determined at any given time. Modes trapped in the surface layer have group speeds on the order of 1435 m/s, compared with 1460 m/s for modes that are not trapped. For a propagation range of 1,000 km, this corresponds to a 12-second delay in arrival time. Clearly, any observations of such large changes an the arrival time of mode 1, 2, or 3 could be the result of capture or release from the surface duct and also provide an estimate of the thermocline depth. The relationship of these travel time changes with other changes needs to be quantified.

Performers: UAF, Shirshov

The fundamental objective of this proposal's modeling component is to develop appropriate methodology for determining the ocean's present climate, and for predicting probable future climate using acoustic data acquired from the Lincoln Sea array. Acoustic transmission data will provide a history of snapshot measurements of the state of the ocean along the slice between the source and the receiver. The first goal will be to determine the present state of the ocean by comparing the real acoustic data with simulated acoustic data from runs with our 3-D coupled ice-ocean model as described below. The second objective is to gain the necessary expertise to begin numerical experiments in acoustic data assimilation. Because acquisition of acoustic data is a relatively recent accomplishment, assimilation of acoustic data into numerical models is in the early developmental stage. Our 3-D coupled ice-ocean model closely simulates the real ocean based on comparisons with hydrographic data, satellite observations, and buoy drift data [Polyakov et al. 1999]. It is now appropriate to begin training in the techniques and problems of incorporating acoustic data into an existing, validated, ice-ocean model.

Currently available atlases, manuals, and reference books [Gorshkov 1980; Levitus 1982; Treshnikov 1985; EWG 1997] include information about multiyear mean atmospheric, oceanic, and ice characteristics, or at best, information about the multiyear mean seasonal cycle. These sources of information do not account for the arctic seasonal evolution related to different arctic climatic states that have been demonstrated in recent studies [Mysak et al. 1990; Proshutinsky and Johnson 1997; Mysak and Venegas 1998; Thompson and Wallace 1998]. For example, Proshutinsky and Johnson [1997] documented two wind-driven circulation regimes, or two climate states, of the arctic atmosphere and ocean, showing that wind-driven ice motion and upper ocean circulation alternate between anticyclonic and cyclonic states. Shifts between regimes occur at 5- to 7-year intervals, resulting in a 10- to 15-year period. The anticyclonic circulation regime (ACCR) has been observed in the model results for 1946–1952, 1958–1962, 1972–1979, and 1984–1988. The cyclonic circulation regime (CCR) prevailed during 1953–1957, 1963–1971, 1980–1983, and 1989–1997. The river runoff, permafrost temperature, ice extent, index of the North Atlantic Oscillation, water temperature in the Iceland Sea, sea ice anomalies in Davis Strait, and other parameters have a similar variability and correlate well with the circulation regimes (see Fig. 18 in Proshutinsky and Johnson 1997). Based on our results to date, we expect substantial differences among the seasonal cycles of ice, oceanic, and atmospheric parameters associated with different arctic climatic states. One goal of this proposed research is to be able to measure and predict shifts between these states.

The seasonal variability of the atmosphere, ocean, and ice in high latitudes depends on coupled dynamics and thermodynamics in ways not yet fully understood. Changes of amplitude or timing of the arctic seasonal cycle can be modulated by interannual variability, implying an important coupling between the seasonal and interannual time scales. The true climatic impact of the seasonal signal on arctic interannual variability remains unknown. Unresolved issues concerning the evolution of the arctic seasonal cycle require further investigation to advance our understanding of climate variability in the region.

This research is designed to improve understanding of the mechanisms affecting the seasonal cycle and the relationship between seasonal and interannual variability, processes that have not been previously stressed in observational studies. Based on Proshutinsky and Johnson [1997] results, we recognize differences among seasonal cycles of sea ice, oceanic, and atmospheric parameters associated with the ACCR and CCR. The acoustic results will be used to establish the differences between the ACCR and CCR seasonal signals and to evaluate a contribution of seasonal changes to Arctic Ocean interannual variability.

Performers: IAP, Shirshov

This task/research are will focus on developing the next generation autonomous acoustic source and mooring design. This effort will build upon the successful design of the ACOUS source deployed in the Franz Victoria Strait in October 1998 (Fig. 3). The entire source complex is shown in Fig. 3, except for the mooring cable and anchor. The source is self-contained with all of the electronics and four of the six battery packs installed inside the radiating complex (two of the battery packs are installed in pressure-proof containers on the outside, one is visible in Fig. 3). This source is controlled by a Rubidium time standard and has a design life of three years. Additional efficiencies in the existing design have already been identified and next generation will be designed to include greater frequency capability and longer life.

This section provides a vision and an overall goal for an integrated Arctic Ocean Observing System building on the concept of acoustic remote sensing and autonomous and cabled moorings. It also shows where the proposed effort would ultimately lead. A future notional Arctic system network building on the ATAM concept with six receive array/ocean observatory cabled moorings and three autonomous acoustic source moorings is shown in Fig. 4. While this network is for illustrative purposes only, it shows that an 18 acoustic path system will sample every major Arctic Ocean basin and intersect every major Arctic Ocean current and frontal structure. The acoustic signal takes 30 mins to propagate along the longest path from Franz Josef Land to the Beaufort Sea array. The ACOUS source currently operating in the Arctic sends a 20.7-min signal every four days. Thus the notional grid shown could provide a snapshot of the entire Arctic Ocean in less than one hour every four days, and could operate unattended for years with all of the data being provided to researchers in real-time. This is very high temporal resolution. As discussed under ATAM Task/Research Area (1) above the Lincoln Sea data would yield direct information as a function of depth, through inversion of the mode arrivals, yielding the temperature changes in the upper layers, the Atlantic layer and the deep Arctic water. Clearly, a tomographic inversion of the notional grid shown in Fig. 4 would yield a horizontal resolution that would vary from 100 kms to many 100s of kms depending upon the density of the crossing paths at any given location. Assimilating the acoustic data and the point measurements continuously into a couple atmosphere-ice-ocean model as described in Task/Research Area (4) above is a key element of the proposed research. Ultimately, resolutions to model scales could be achievable. The current 3D model described in ATAM Task/Research Area (4) above is being run on a 55.56-km resolution and 17 km is the next step.

The Acoustic Thermometry and Autonomous Monitoring (ATAM) installations would be designed to be as simple and cost effective as possible exploiting Arctic Ocean acoustic propagation physics. As discussed in task/research area (1) above the Arctic Ocean acoustic waveguide is highly dispersive. This means that each mode travels at a different group speed. The longer the acoustic path the more time spread there is between modal arrivals. Separating the modes in time in principal requires only one hydrophone and not an array, and that could realize significant cost savings with very simple receive systems. There is a tradeoff between source bandwidth (which sharpens the modal arrival peaks in time requiring less time spread to separate the peaks) and path length that would need to be done to develop the required source bandwidth. Preliminary calculations indicate that a bandwidth of 5 Hz would be sufficient for the shorter paths (the currently operating ACOUS source has a bandwidth of 2 Hz, 5 Hz is easily achievable). The network ATAM moorings would include thermisters, salinometers, current meters, chemical and biological sensor packages, and seismometers. Analyses would be done to see if upward looking sonar and ADCPs can be included depending upon power budgets and trunk cable power capabilities. Data and power reserve would be built in to accommodate emerging technologies such as AUV support.

The national network shown in Fig. 4 represents approximately 24,000 kms of acoustic path length. In order to obtain an average temperature on these 18 acoustic paths that would be equivalent to the acoustic thermometry, by direct measurement of temperature, it would require sampling along each path at the mesoscale correlation length of approximately 25–50 kms (for the SCICEX cruises the CTD sample spacing has been an average of 40 kms). This would therefore require installing and maintaining a continuous presence of 600–1,000 buoys equipped with at least a 1 km long thermister chains that must be mounted through the ice in locations that span the entire Arctic, which given the realities of Arctic logistics is operationally as well as economically unfeasible. The International Arctic Buoy Program (IABP) maintains 20–30 buoys at any given time on the ice that measure atmospheric variables and are tracked for ice drift. Their goal is to maintain a spacing not less than 500 km. This is two orders of magnitude less than required and the buoys that would measure ocean temperature or acoustic data would be much more difficult to transport and deploy with their long arrays. However, the Arctic buoys are the right technology for sampling the atmospheric properties and ice drift and will be a critical part of an integrated Arctic observing system when coupled with the Arctic Ocean observing system. Much smaller numbers of oceanographic and acoustic buoys would be feasible for ice and thermocline acoustic measurements as discussed in ATAM Task/Research Area (2) above.

A preliminary cost analysis of the national system depicted in Fig. 4 has been performed. The network shown in Fig. 4 includes two seashore interfaces, one in Barrow, Alaska, and one in Alert, Canada. The seashore termination is where the cable crosses shore and must be protected from shore-fast and rafted ice. The seashore termination already exists in Alert, Canada, and is a slant-drilled conduit that goes from the shore to a point at sea well beyond any potential coastal ice-scour zones. The sea-shore interface at Barrow, would involve jetting in reinforced drill pipe to a depth of 2–3 meters or slant drilling which would be even more robust, but would be approximately $500,000 more than jetting in for a total of approximately $1 million for the sea-shore interface. The latter cost is used in the preliminary cost estimate below.

The total cost is $26 million, and for a 5-year program of observations amortized costs are $5.1 million/year, for a 10-year program (replacing sources and ATAM arrays once) amortized costs are $3.2 million/year. An additional $1 million/year is likely needed to support the research and operations costs. For a 10-year program this yields a total of $42 million or $4.2 million /year. For comparison the SHEBA program cost to date is approximately $22 million with an estimated $3 million /year for three years to support the research and data analysis, for a total of $31 million. The SCICEX submarine program has been running at approximately $2–3 million /year (exclusive of the acquisition and operating costs of a nuclear submarine!) for a total cost of approximately $20 million if a SCICEX 2000 is executed. The largest cost of the ATAM/ACOUS network is the acquisition of the undersea trunk cable. Termination to Russian coasts could cut this cost in half, substantially reducing the cost of the network. Additional cost reductions in the ATAM arrays are possible with much simplified acoustic receivers exploiting the modal time dispersion discussed above. There is no other technology that can provide a decade of synoptic, real-time, year-round, distributed sampling of the Arctic Ocean with costs that are comparable to contemporary Arctic initiatives.

Fig. 4. A future national monitoring grid in the Arctic Ocean building on the ATAM concept and utilizing the existing seashore termination from Alert, Canada, into the Lincoln Sea to realize a cabled network of Acoustic Thermometry and Autonomous Moorings (ATAM) for real-time, year-round observation in the Arctic Ocean.

The data collected by the ACOUS Lincoln Sea Array that will be recovered in October 2000 will be made available to other interested scientists and researchers on an anonymous ftp site at the University of Alaska Fairbanks. The availability of the data will be published on appropriate web sites such as: NOAA, IARC, and CIFAR.

The data will be archived at the National Snow and Ice Data Center (NSIDC), Boulder, Colorado. NSIDC is an information and referral center supporting polar and cryospheric research. NSIDC distributes snow and ice data, and maintains data archives about snow cover, avalanches, glaciers, ice sheets, freshwater ice, sea ice, ground ice, permafrost, atmospheric ice, paleoglaciology, and ice cores. The index of NSIDC data is located at www-nsidc.colorado.edu/NSIDC/CATALOG/index.html.

Bjorgo, E., O.M. Johannessen, and M.W. Miles, "Analysis of merged SMMR-SMMI time series of Arctic and Antarctic sea ice parameters 1978-1995," Geophys. Res. Lett., 1997.

Carmack, E.C., R.W. MacDonald, R. G. Perkin, F.A. McLaughlin, and R.J. Pearson, "Evidence for warming of Atlantic Water in the southern Canadian Basin in the Arctic Ocean: Results form the Larson-93 expedition," Geophys. Res. Letters., v.22, 9, pp. 1061-1064, 1995a.

Carmack, E.C., K. Aagaard, J.G. Swift, R.W. Macdonald, F. A. McLaughlin, R. G. Perkin, and P. E. Jones, "The Arctic is warming: Results from the US/Canada 1994 Arctic Ocean Section," In: The Abstracts of the XXI General Assembly of IAPSO, Honolulu, Hawaii, pp. 26-27, 1995b.

Carmack, E.C. and K. Aagaard, "The dynamic Arctic Ocean: Spatial, Temporal, and Conceptual Heterogeneities," Abstract, EOS, Transactions, AGU, Vol. 76, Nol 3, OS12, 1996.

Environmental Working Group (EWG), "Joint U.S.-Russian Atlas of the Arctic Ocean" [CD-ROM], Natl. Snow and Ice Center, Boulder, Colorado, 1997.

Gavrilov, A.N., P.N. Mikhalevsky, and M.Yu Andreyev, "Measurements of low frequency transmission loss in the Trans-Arctic Acoustic Propagation Experiment," Proceedings Oceans 95, MTS/IEEE, Vol. 2, pp. 941-948, San Diego, CA., 1995.

Gavrilov, A.N. and P.N. Mikhalevsky, "Modeling an acoustic response to long-term variations of water and ice chracteristics in the Arctic Ocean," Proceedings Oceans 95, MTS/IEEE, Vol. 1, pp. 247-253, San Diego, CA., 1995

Gavrilov, A.N. and P.N. Mikhalevsky, "Mode coupling effects in acoustic thermometry of the Arctic Ocean," submitted (in review), J. Acous. Soc. Am., 1999.

Gavrilov, A.N. and P.N. Mikhalevsky, "Phase Acoustic Tomography with Broadband Signals in the Arctic," Proceedings 3^rd European Conference on Acoustics, Crete, Greece, June 1996.

Gorshkov, S.G. (ed.), Atlas of the Oceans, Arctic Ocean, Leningrad, GUNIO(in Russian), 1980.

Grotenfendt, K., K. Logemann, D. Quadfasel, and S. Ronski, "Is the Arctic Ocean warming?", J. Geophys. Res., Vol. 103(C12), pp. 27,679-27,687, 1998.

Hakkinen, S. and G.L. Mellor, "Modeling the seasonal variability of a coupled Arctic ice-ocean system," J. Geophys. Res., 97.20285-20304, 1992.

Hibler, W.D. III and K. Bryan, "A diagnostic ice-ocean model," JK. Phys. Oceanogr., 17, 987-1015, 1987.

Johannessen, O.M., M.W. Miles, and E. Bjorgo, "The Arctic's shrinking sea ice," Nature, 376, 126-127,1995.

Johannessen, O.M., E. Bjorgo, and M.W. Miles, "Global Warming in the Arctic," Science, Vol. 271, pp. 1219, 12 January 1996.

Johnson, M., A. Proshutinsky, and I. Polyakov, "Atmospheric Patterns Forcing Two Regimes of Arctic Ice-Ocean Circulation: A Return to Anticyclonic Conditions?," in review, Geophys. Res. Let., 1999.

LePage, K. and H. Schmidt, "Modeling of low-frequency transmission loss in the central Arctic," J. Acoust. Soc. Am., Vol. 96, No. 3, pp. 1783-1795, 1994.

Levitus. S., "Climatological atlas of the world ocean," NOAA Publ., 13, 173 pp., U.S. Dept. of Commerce, Washington, D.C., 1982.

Keenan, R.E., "Ice Surface Loss Models Evaluation and Recommendation," SAIC-89/1456, 1989.

McLaughlin, F., E.C. Carmack, R.W. MacDonald, and J.K. Bishop, "Physical and geochemical properties across the Atlantic/Pacific water mass front in the southern Canadian Basin," J. Geophys. Res., Vol. 101, pp. 1183-1197, 1996.

Mikhalevsky, P.N., "Characteristics of CW signals propagated under the ice in the Arctic," J. Acoust. Soc. Am., Vol 70, 1717-1722, 1981.

Mikhalevsky, P.N., R.D. Muench, and F.R. Dinapoli, "Arctic Ocean Warming: Can We Measure it Acoustically?," The Oceanographic Society, Second Scientific Meeting, St. Petersburg, FL, Mar 1991.

Mikhalevsky, P.N., A.B. Baggeroer, A.N. Gavrilov, and M. Slavinsky, "Experimental test use of acoustics to monitor temperature and ice in the Arctic Ocean," Eos , Vol. 76, pp.27, 1995a.

Mikhalevsky, P.N., A. Gavrilov, and A.B. Baggeroer, "Are faster than predicted signals seeing Arctic Ocean warming?," J. Acoust. Soc. Am., Vol. 97(5), Pt. 2, May 1995b.

Mikhalevsky, P.N., R.E. Keenan, and A.B. Baggeroer, "Measured transarctic acoustic travel times and model comparisons," in Proceedings of 3^rd European Conference on Underwater Acoustics, 242-28 June 1996.

Mikhalevsky, P.N., A.N. Gavrilov, and A.B. Baggeroer, "The Transarctic Acoustic Propagation Experiment and climate monitoring in the Arctic," in press, IEEE J. Oceanic Engineering, 1999.

Morison, J., M. Steele, and R. Anderson, "Hydrography of the Upper Arctic measured from the nuclear submarine USS Pargo," submitted to Deep Sea Research, 1996a.

Morison, J., M. Steele, and R. Anderson, "Changes in Upper Ocean hydrography measured during the 1993 cruise of the USS Pargo," Abstract, Eos, Transactions, AGU, Vol 76, No. 3, OS12, 1996b.

Mysak, L.A., D.K. Manak, and R.F. Marsden, "Sea-ice anomalies observed in the Greenland and Labrador seas during 1901-1984 and their relation to an interdecadal Arctic cycle," Clim. Dyn.,5,111-113, 1990.

Mysak, L.A. and S.A. Venagas, "Decadal climate oscillations in the Arctic: A new feedback loop for atmospheric-ice-ocean interactions," Geophys. Res. Letters, 25,19,3607-3610, 1998.

Nikiforov, J.G. and A.O. Shpaikher, "Features of the formation of hydrological regime large-scale variations in the Arctic Ocean," 269 pp., Gydrometeoizdat, Leningrad 1980, (in Russian).

Oberhuber, J.M., "Simulation of the Atlantic Circulation with a coupled sea ice-mixed layer-isopycnal general circulation model. Part 2: Model Experiment," J. Phys. Oceanogr., 23, 830-845, 1993.

Overpeck, J., K. Hughen, D. Hardy, R. Bradley, R. Case, M. Douglas, B. Finney, K. Gajewski, G. Jacoby, A. Jennings, S. Lamoureux, A. Lasca, G. MacDonald, J. Moore, M. Retelle, S. Smith, A. Wolfe, and G. Zielinski, "Arctic Environmental Change of the Last Four Centuries," Science, Vol. 278, pp 1251, November 1997.

Piacsek, S., R. Allard, and A. Warn-Varnas, "A study of the arctic ice cover and upper ocean with a coupled ice-ocean model," J. Geophys. Res., 96, 4631-4650, 1991.

Pawlowicz, R., D. Farmer, B. Sotirin, and S. Ozard, "Shallow-water receptions from the Transarctic Acoustic Propagation Experiment," J. Acoust. Soc. Am., 1996.

Polyakov, I., A. Proshutinsky, and M. Johnson, "Two seasonal cycles of the Arctic Ocean from modeling and observations", to be submitted, J. Geophys. Res., 1999.

Proshutinsky, A.Y. and M.A. Johnson.,"Two circulation regimes of the wind-driven Arctic Ocean". J. Geophys. Res., 102(C6), 12493—12514, 1997.

Semtner, A.J., "A numerical study of sea-ice and ocean circulation in the Arctic," J. Phys. Oceanogr., 17, 1077-1099, 1987.

Steele, M. and T. Boyd, "Retreat of the cold halocline layer in the Arctic Ocean," J. Geophys. Res., 103(55), 10,419-10,435, 1998.

Thompson, D.W.J. and J.M. Wallace, "The Arctic Oscillation signature in the wintertime geopotential height and temperature fields," Geophys. Res. Lett., 25, 1297-1300, 1998.

Treshnikov, A.F., "Water masses of the Arctic Ocean," in Polar Oceans, M.J. Dunbar (ed.), 1977.

Treshnikov, A.F., (ed.), "Atlas of the Arctic," Moscow, GUGK (in Russian), 1985.

Vinnikov, K.Y., A. Robock, R.J. Stouffer, J.E. Walsh, C.L. Parkinson, D.J. Cavalieri, J.F.B. Mitchell, D. Garrett and V.F. Zakharov, "On the Detection and Attribution of Anthropogenic Global Warming Using Northern Hemisphere Sea Ice Extent," submitted to Science, August 1999.

Dr. Mark Johnson, University of Alaska Fairbanks - Dr. Johnson will coordinate the overall activities of the Acoustic Thermometry and Autonomous Monitoring effort. He will be the principal point of contact with IARC and CIFAR. In addition, Dr. Johnson will lead the work to assimilate the data from the ACOUS receive array into the UAF atmosphere-ice-ocean models.

Dr. Peter Mikhalevsky, Science Applications International Corporation - Dr. Mikhalevsky will coordinate the activities of SAIC in the ATAM effort. Dr. Mikhalevsky will lead the analysis of the data from the ACOUS Lincoln Sea Array and the data collected in the Chukchi Sea in April 1999. He will also lead the acoustic remote sensing research effort. Dr. Mikhalevsky serves on several international Arctic and Ocean advisory committees. He will insure that the ATAM effort for IARC and CIFAR will be recognized in international scientific forums.

Dr. Arthur Baggeroer, Massachusetts Institute of Technology - Dr. Baggeroer will coordinate the activities of MIT in the ATAM effort. He will lead MIT's effort in theoretical modeling and analysis to quantify the techniques for ice intensity tomography, and support the signal and array processing of the data for travel time estimation and ice intensity tomographic inversion.

Prof. Konstantin Naugolnykh, and Prof. Er-Chang Shang, Environmental Technology Laboratory, NOAA - Prof. Naugolnykh and Prof. Shang will coordinate the activities of ETL in the ATAM effort. They will lead ETL's effort in analyzing the effect of temperature and current change on acoustic propagation in the Arctic Basin, and analyzing the effect of sea ice thickness and roughness on Transarctic sound signal propagation.

Dr. Alexander Gavrilov, Shirshov Institute of Oceanology, Moscow, Russia - Dr. Gavrilov will coordinate the activities at the Shirshov Institute in the ATAM effort. He will lead the effort to perform beamforming, modal detection, and pulse compression processing and phase detection to determine the arrival times of acoustic pulses as a function of acoustic mode. He will also evaluate inversion methods using modal travel times and make recommendations about future planned acoustic sources and receive arrays.

Dr. Boris Bogolyubov, Institute of Applied Physics, Nizhny Novgorod, Russia - Dr. Bogolyubov will coordinate the activities at IAP in the ATAM effort. He will lead the design of the next generation autonomous acoustic source and mooring to be installed in the Central Arctic Basin. His team has already identified efficiencies in the design of the existing source to include greater frequency capability and longer life.