Final Report Summary - ACOBAR (Acoustic Technology for observing the interior of the Arctic Ocean)
Executive Summary:
The overall objective of ACOBAR has been to develop a system for environmental monitoring of the interior of the Arctic Ocean by use of acoustical methods, in situ point measurements and numerical models including data assimilation. During annual field experiments fro 2008 to 2012, ACOBAR has demonstrated several acoustic methods, including tomography, data transmission and communication to/from underwater platforms, and navigation of gliders. The ACOBAR results provide complementary methods to the ARGO system, which cannot be used in ice-covered seas, based on platforms located under the sea ice. Data collection and transmission from the water column and the seafloor has been tested using acoustics methods on moorings, gliders and ice-tethered platforms. ACOBAR has developed methods that can be used in the polar regions, contributing to filling gaps in the global ocean observing system. ACOBAR has conducted field experiments with acoustic sources and receivers, as well as gliders in the Fram Strait and drifting ice-tethered platforms in the high Arctic Ocean. An array of acoustic sources and receivers was operated over two years for measurement of acoustic travel time, and methods have been developed to retrieve integrated ocean temperature from the travel time data. Acoustic tomography in combination with in situ data and numerical models is an innovative method to obtain integrated 3-D fields of temperature, transports and heat fluxes in the ocean. Long-range acoustic signal propagation has been used to develop localisation methods for gliders when they are in sub-surface position. These methods will be the basis for acoustic navigation of gliders, which will be required for gliders to operate under the ice-cover. Also data transmission from fixed moorings via acoustic modems to the surface for downloading from ships or for satellite transmission has been tested and useful experience has been gained for further development of underwater data transmission. The acoustic technologies aim to be used for transmission of multidisciplinary data from underwater observatories, which are presently under development in many sea areas. Transfer of technology and know-how between USA and Europe and between industry and research institutes has been an important component of the project. The exchange of personnel between the partners, workshops and meetings between scientists, engineers and students have made it possible to build up expertise for planning and implementation of complex field experiments and obtain unique data sets. The first results of the project have been published in referee journals and popular science articles as well as presentations at conferences and workshops. Further publications and other dissemination of results will be done after the project is completed.
Project Context and Objectives:
The Arctic Ocean is becoming increasing important as a result of the global warming and the growing economical and political interest in the region. Sea ice reduction is facilitating resource exploration, marine transport and other economic activities in the Arctic and sub-Arctic regions. It is anticipated that over the coming decades the Arctic is likely to attract substantial investment, potentially reaching $100bn or more. It is therefore important to improve our understanding about the climatic and environmental changes in the region, along with their socioeconomic impact. On this background the Arctic is an area of growing strategic importance for Europe, as shown in a number of policy documents from 2008 to present (http:// http://eeas.europa.eu/arctic_region/). EU’s Arctic policy is focussed on three policy areas: (1) protecting and preserving the Arctic in unison with its population; (2) promoting the sustainable use of resources; and (3) international cooperation.
The Arctic Ocean is poorly observed and monitored compared to the other world oceans. It is clear that the issues associated with Arctic marine change transcend national boundaries and science disciplines. No one nation has the sovereignty, expertise, knowledge, logistical capability or budget to individually tackle these challenges head on; a truly international and integrated scientific effort is needed. Understand Arctic marine change needs a technological, multidisciplinary, and integrated approach. While satellite observations are important for monitoring the ocean surface including sea ice, there is severe lack of in situ and subsurface data for the Arctic Ocean. Such information is essential for research on climate, marine resources, shipping (tourist and commercial), oil and gas exploration, support to operations and protection of the environment. It is therefore of high priority to create and sustain an Arctic Ocean observing system that can provide services to different users groups such as policy makers, climate researchers, marine scientists, environmental agencies, management of natural resources, shipping and other maritime activities. At present collection of in situ ocean data in the Arctic seas is generally performed by icebreakers, aircraft and submarine expeditions, which are all very expensive and not suitable for providing regular, long term monitoring data. A buoy programme with Ice-Tethered Profilers has started, but covers only a limited part of the Arctic Ocean (http://www.whoi.edu/page.do?pid=20756).
In recent decades, the Arctic Ocean has experienced dramatic changes documented by the record low summer minimum ice extent (http://arctic-roos.org) and the strong reduction of multiyear ice (The Economist June, 2012). Atmospheric warming is a dominant forcing for the melting of ice, but evidence is now building that warmer ocean also contributes to the thinning of Arctic ice (Carmack and Melling, 2011). It is well documented that the ice is getting thinner and more seasonal (Wadhams, 2012), and this reduction in ice coverage will lead to stronger exposure of the surface layer to wind-induced mixing. In the Arctic Ocean a 100-200 m thick surface layer of low salinity and cold water protects the ice cover from more extensive melting by the warmer underlying Atlantic water. Under intensified wind forcing the isolating surface layer can easily destroyed due to mixing with warmer water masses and this process may accelerate the sea ice melt beyond what we have seen so far. The accelerated melting observed during the last years might indicate that this process already is in progress. The heat accumulated in the ocean can also retard the ice growth through the fall and into the winter and this process can be substantially impacted by upper ocean stratification changes (Steele et al, 2010). However, the currently available climate models are yet not able to reproduce the observed accelerated ice melting (Rampal et al. 2011). The weak turbulent mixing in the upper layer and consequent small vertical heat flux allow the existence of a cold halocline in the Arctic Ocean (Fer 2009). However, most of turbulence in the Arctic region is associated with internal waves over deep topography while near-inertial waves, generated by ice pack motions also play an important role for the upper ocean mixing. Recent findings show that increased near-inertial waves generated in the presence of ice pack at the surface will enhance shear and mixing in the Arctic (Rainville and Woodgate 2009). A seasonally ice free Arctic Ocean can evolve towards conditions similar to in the marginal ice zones, and this can lead to a redistribution of water masses, meaning that the isolating halocline layer can be weakened and vertical heat flux from the underlying Atlantic water will increase.
Since the 1990s, in particular during the International Polar Year (2007-2009), ocean observations aimed in monitoring the volume, heat and freshwater fluxes between the Arctic Ocean and the subpolar seas have been significantly enhanced. However, when using available estimates of long-term mean fluxes, the mass and heat budget of the Arctic Ocean still cannot be balanced, mainly because the uncertainties in the Fram Strait fluxes are larger than the total fluxes through the other gateways (Beszczynska-Möller et al., 2011). Fram Strait is the main deep passage through which the ocean mass and heat exchange between the Atlantic and Arctic Ocean takes place. On the eastern side of the strait the northbound West Spitsbergen Current (WSC) transports Atlantic water to the Arctic Ocean, whereas on the western side the East Greenland Current (EGC) transports sea ice and polar water from the Arctic Ocean to the south, towards the Nordic Seas and Atlantic Ocean. The topographic structure of the strait causes a splitting of WSC into at least three branches, of which one recirculates between 78°N-80°N.
An oceanographic moored array has operated since 1997 to monitor the ocean water column properties and oceanic advective fluxes through Fram Strait. An extremely complex circulation pattern has been observed, characterized by a pronounced recirculation and strong mesoscale activity. Since the spatial resolution of the mooring array varies from 10 to 30 km, it is not sufficient for resolving the mesoscale variability and recirculation, and large errors exist in estimates of the oceanic fluxes. At present the moored observing system does not have any surface expression and therefore cannot perform measurements of biogeochemical and hydrographic parameters in the upper water column (0-50 m). Moreover, below 1000m a vertical resolution of measurements is very low which can introduce a significant bias in the vertical stratification and circulation in deep layers. An improved observing system is required, with high temporal and spatial resolutions capable of quantifying the impact of mesoscale currents and determining the influence of the recirculation on net oceanic transports through the strait. Such an integrated observing and model system for long-term environmental monitoring in Fram Strait, combining data from satellites, acoustic systems, moorings, and gliders with high-resolution ice-ocean circulation models through data assimilation has been under development in ACOBAR project (Sagen et al., 2010, Sandven et al. 2012).
The main objective of ACOBAR has been to develop observing systems for environmental monitoring of the deep Arctic Ocean by assimilation of data obtained with acoustical methods including tomography/thermometry, data transmission from underwater platforms, communication and navigation of floats and gliders under the ice-cover. The observing system has been developed and tested in two different areas: the Fram Strait, which is partly ice-covered and partly ice-free, and the high Arctic which is fully ice-covered throughout the year. The Fram Strait observatory consists of moored acoustic sources and receiver for tomography, and gliders combined with ship surveys, moored profilers and satellite observations of sea ice. In the high Arctic, also ice-tethered platforms were used to observe changes of water masses. Improved observation of the heat, mass and freshwater transport in the whole Arctic is important for climate research.
The specific objectives have been:
1) Develop improved 3-D observations of the properties and transport of water masses in Fram Strait using an acoustic tomography array, oceanographic moorings and profiling gliders through data assimilation.
2) Test and assess transmission of long-range acoustic navigation signals and commands to operate autonomous underwater vehicles such as gliders and floats in the Fram Strait.
3) Implement data transmission by acoustic modems from underwater platforms to the surface for downloading to ships or satellite transfer.
4) Exploit the existing array of acoustic sources from ice-tethered platforms for tomographic measurements of water mass properties in the ice-covered Arctic Ocean with real-time data provision via satellite.
5) Dissemination of underwater oceanographic data to users with near real-time capability.
6) Provide the technology to combine the oceanography array to transmit data through Arctic cabled observatories when available.
7) Transfer of technology and know-how between USA and Europe, and between industry and research institutes.
Project Results:
Results from WP1: Design, synthesis and assessment
The acoustic tomography experiment was carried out from 2010 to 2012 in a triangle of transceivers moorings with a moored receiver in the center. This geometry gave a total of six acoustic paths that covered the Fram Strait Marginal Ice Zone. The acoustic transceivers also transmitted RAFOS signals to provide an underwater acoustic navigation system for gliders and floats. Several glider experiments were conducted where the acoustic signals from the sources were used to develop acoustic navigation under ice. This experiment was the first implementation of a multipurpose acoustic network for tomography, passive acoustic monitoring, and navigation of gliders. The primary goal is to improve the accuracy of the heat, mass and freshwater transport estimates through the Fram Strait using a four-dimensional data and model system combining acoustic travel time measurements and ocean data from gliders and moorings with high resolution ice-ocean modelling through data assimilation. Another goal is to observe and explain changes in ambient noise with respect to changes in environmental conditions and increased human activities (Sandven et al, 2012).
The ACOBAR project has successfully shown that network of acoustic transceivers and receivers can be used as a infrastructures for acoustic tomography covering the whole Arctic basin, underwater GPS system for tracking of floats, assistance to navigation of underwater vehicles, and for passive monitoring of marine mammals, human activity and environmental conditions (ice dynamics, wave conditions). This makes the acoustic networks to an important component in a future integrated Arctic observing system (e.g. Sagen et al. 2010, Lee et al. 2010). The interdisciplinary benefit of acoustic components in an integrated Acoustic Observing system is described Community White paper submitted by Mikhalevsky, Sagen, Worecester et al. 2013 to the first Arctic Observing Summit. This paper demonstrates the capability and interest in using acoustics as part of an integrative observing system in the Arctic. The community paper suggests the following recommendations for implementation of a future observing system:
1) A multiyear development and implementation plan of multipurpose acoustic network infrastructure that includes international participation should be completed. This plan should start with one or two cabled nodes like the National Science Foundation Ocean Observatories Initiative (OOI) (http://www.oceanobservatories.org/) Regional Scale Nodes (RSN), and Canadian Neptune (http://www.neptunecanada.com/) perhaps off the North Slope of Alaska into the Beaufort Sea or coordinated with the Svalbard Integrated Observing System from the Fram Strait and eventually into the Eastern Arctic.
2) Further develop reliable low- to mid-frequency sources (50-500 Hz) to support the basin and regional scale tomographic, communications and navigation networks. Multipurpose network of regional sources and receivers on moored and ice-tethered platforms should be deployed and maintained annually. These deployments should be coordinated with icebreaker expeditions and other logistical capacities that are expected to be available every year in the high Arctic.
3) Transfer the technology products and lessons learned on the subsea infrastructure being deployed for the OOI RSN into the Arctic.
4) Integration of acoustic data with oceanographic data from ice-tethered platforms, gliders and ships as well as sea ice data, where spatially integrated measurements from acoustics are used in combination with point measurements and vertical profiles of physical variables. These data can then be used in modelling and data assimilation systems.
The implementation of a sustainable Arctic Observing system, including the acoustic networks will need international funding, coordination and roadmaps. On the European side several roadmap project are planned such as the SIOS- The Svalbard Integrated Earth Observing System (Cynan Ellis-Evans, et al. 2013, http://www.sios-svalbard.org ) and the EPOS - European Plate Observing System (see Kuuvet et al. 2013, http://www.epos-eu.org/). The ACOBAR consortium will integrate and collaborate with those projects as well as with bodies such as EuroGoos.
An environmental assessment study was conducted in the beginning of the project as part of the design study, and it was updated by the end of the project. The results were reported in two Environmental Assessment Reports (EAR1 and EAR2). The field experiments built upon the initial underwater acoustic tomography system installed in Fram Strait during 2008 by the DAMOCLES Integrated Project. A post-experiment acoustic and environmental analysis of sound transmissions has been conducted to confirm the predicted potential for impacts on the marine environment that was documented in EAR1. Based on the actual implementation of the experimental equipment and scientific analysis of the acoustic sources, the experiment did not impact the environment of Fram Strait. The low source level of the acoustic sources, combined with their placement in relationship to Marine Protected Areas (MPAs) and the long intervals between transmissions, precluded the experiment activities from affecting any resources of an MPA. Based on the potential for impacts, it is highly unlikely that the experiment activities affected any marine species listed as endangered or threatened under the U.S. Endangered Species Act nor any marine species that are listed as threatened (vulnerable, endangered, or critically endangered) by the International Union for Conservation of Nature (IUCN). Additionally, is very unlikely that any behavioral takes of marine mammals (under the U.S. Marine Mammal Protection Act) occurred as a result of this experiment. In conclusion, the scientific analysis of the ACOBAR experimental activities indicated that no ethical issues resulted, as the experiment had no potential for causing grave danger to marine mammals potentially occurring in the Fram Strait area.
Results from WP2: Arctic Ocean Observatory
The main activity in WP2 has been to develop, test and implement a system of Acoustic Ice Tethered Platforms (AITPs) which can operate over long time in Arctic ice covers areas. The AITP system is a cluster composed of 4 distinct but identical platforms used for navigating and exchanging data with an acoustic under sea-ice glider. The acoustic sea-ice glider is not a conventional sea glider since it is deprived of any direct links with satellites except for an Argos antenna used in case of emergency and back up for ultimate retrieval. The AITP platforms are fully autonomous over periods of several months. Each platform is composed of a surface unit equipped with a satellite Iridium and a GPS transmitter, an underwater ocean profiler moving along a 1000m long cable suspended vertically under the surface buoy, three underwater acoustic modems for exchanging data between the sea-ice glider, the ocean vertical profiler and the surface unit.
In addition, the ocean vertical profiler is equipped with a CTD sensor for measuring temperature and salinity versus depth from surface down to 1000m depth and a SOFAR-RAFOS low frequency acoustic transmitter/receiver (1560-780hz) for transmitting long range acoustic signals to the sea-ice glider. Each AITP is capable of transmitting and receiving SOFAR-RAFOS acoustic signals several times per day in order to estimate precisely the sound speed between AITPs and the sea-ice glider for navigation and localisation purposes. The underwater vertical ocean profiler is equipped with an hydraulic buoyancy control device composed of two bladers filled up with oil in order to displace the ocean profiler up and down along the 1000m long cable suspended under the surface unit. The ACOBAR AITP cluster is composed of 4 identical platforms operating in a similar way. Each Ocean profiler is profiling once per day from surface down to 800m depth to measure temperature, electrical conductivity and pressure from which one can deduce potential temperature, salinity, density and depth. This profile is transmitted from the ocean profiler to the surface unit via the acoustic modems installed both on the ocean profiler and the surface unit. Then this vertical CTD profile is transmitted via Iridium to satellites in near real time. CTD profiles are then stored on a server at the French Polar Institute (IPEV).
In addition to the CTD profiles, the ocean profiler is equipped with a Seascan system to send SOFAR and to receive RAFOS signals. The system operates once per day and uses 4 dedicated acoustic windows. For each window only one ocean profiler is in sending mode while the 3 other ocean profilers are in receiving mode. Each ocean profiler is configured to use 25 minute window starting at 11h00 each day. The Seascan RAFOS system returns the TOA of the six highest correlation level signals received during one acoustic window. The Seascan SOFAR system uses a 224 samples sweep at a resolution of 0.09230 second per sample and corresponding to a 20 second sweep.
The AITP is transmitting acoustic data from the SOFAR-RAFOS system (TOA and Correlation levels) and some technical data (voltage, pumping actions etc...). TOA concern Time of Arrival of the acoustic SOFAR signals transmitted from one AITP to the 3 other AITP part of the same cluster. One can then compare the travel time between pairs of AITP with GPS positions in order to deduce underwater sound speed. This is the sound speed that will be used for navigating glider under sea-ice. TOA need to be corrected from ocean profilers clock drift. TOA resolution is about 0.1s (equivalent to about 150m). Levels of correlation are giving an indication about the strength of the acoustic signal received by each AITP ocean profilers. In fact the 6 best correlation levels are recorded inside the same window frame. Only correlation levels well above noise level are selected for each TOA.
Readiness of AITP cluster for testing in Arctic Waters is mainly related to the capacity of AITP to garantee a full time operational behaviour by accomplishing (1) the basic transmission schedule both involving the underwater double acoustic system (SOFAR-RAFOS and Acoustic modems) several times per day, (2) the satellite data transmission schedule (Iridium) and GPS positioning several times per day and (3) the daily vertical ocean profiles from surface down to 1000m depth maximum. In order to test it we selected the transpolar drift terminal section from the North Pole to Fram Strait between Spitsbergen and Greenland. Deployments of AITP occurred during Spring time near the North Pole taking advantage of the Russian Barneo station installed near the Pole every year for one month in April. The recovery was organized during the Fall (September-October) from Norwegian vessels operating in the Fram Strait area (Lance and KV Svalbard). The duration of the test was about 5 to 6 months covering Spring and Summer (winter and summer conditions). Two tests were conducted during ACOBAR in 2012 and 2013 respectively over a 6 month period and allowed to detect and fix several technical problems.
Basically the AITP is now proven for operating in real Arctic conditions from winter (freezing) to summer (melting). There are still some improvements to be done for increasing the duration of operations in particular for ocean profilers to profile for 6 months. The actual limitation is now of about 3 months due to several problems for activating the hydraulic pump. Too many pumping actions for each profilers are decreasing the autonomy of the ocean profiler. The ocean profiler is a large underwater neutrally buoyant float with some drag effect that needs to be compensated and generate more pumping actions increasing energy demand significantly. A single rider instead of two for the actual version of the ocean profiler might be able to decrease the drag along the 1000m long cable during each daily vertical profiles. The SOFAR-RAFOS acoustic system needs to be improved for increasing range of propagation for the acoustic signals allowing a better tracking of the under sea-ice glider. Both the strength of the acoustic SOFAR signals and a better precision and resolution of the RAFOS detection (Correlation) are needed. Finally the Iridium transmission and GPS localisation need also to be improved. Thanks to the addition of an Ice Mass Balance (IMB) system on each AITP measuring temperature profiles across sea-ice and equipped with an autonomous Iridium and GPS transmitter, we were able to realize a good tracking for each AITP during the two operations in 2012 and 2013. It was quite fortunate since we met some problems with AITP iridium-GPS transmitters during the two years.
The Ocean profilers are quite reliable and the quality of the data is excellent. The Acoustic modems are also quite reliable and the data transmission between the ocean profilers and the surface units is quite good. The general and overall configuration of the AITP is quite well adapted to operate in the Arctic during freezing and melting conditions. This is facilitating deployments and recoveries as well. The region selected for these experiments extended from the North Pole down to Fram Strait over more than 1000 km range. The second test in 2013 consisted in two AITP platforms deployed in the vicinity of the North Pole in April 2013 on two ice floes located 10 km apart. It took about 6 months (April to September 2012) to drift over more than 1000 km between the North Pole and Fram Strait.
The data collected consists of about 60 vertical profiles of temperature and salinity obtained from both AITPs over a three months period demonstrating the reliability of the whole system. The data shows the presence of a cold subsurface layer at freezing temperatures located at about 100m depth enriched with brines rejected during sea-ice formation in winter. The resulting salinity for this brines enriched layer was about 34 ppt salinity. At greater depths (300m to 400m) the Atlantic layer is quite recognizable with temperatures exceeding +1°C and salinities reaching 34.80psu salinity. Surface salinities were low at the North Pole (30psu) and were increasing further south (34psu salinity).
The underwater ocean profiler, which also includes a dual acoustic system, provides CTD data from surface down to 800m depth on a daily basis. The CTD data are fundamental for observation of water masses and ocean processes under the sea ice. In addition the CTD profiles can be used for estimating the suitable depth for optimizing long range acoustic propagation between AITPs and the sea glider operating under sea-ice. The dual acoustic system operating between the AITPs is composed of long range SOFAR transmitters and RAFOS receivers and short range acoustic modems for data exchange. Therefore the AITP cluster provides essential information to the sea glider in order to navigate under sea ice over long distance (SOFAR-RAFOS) and to exchange data at short range (Modems). The AITP surface unit provides all the necessary information concerning real time positioning of the AITP using GPS. In addition the surface unit communicates with satellites in order to transmit data and exchange information in both directions between land stations and the AITPs.
From a technological point of view, the ACOBAR AITP is now a proven and reliable instrument based on two consecutive tests occurring in real Arctic conditions covering both winter (freezing) and summer ( melting) seasons over a 6 month period and over more than 1000km range. The acoustic modem transmission between the surface unit and the underwater unit was quite reliable. The ocean profilers have completed between 40 and 60 high quality vertical CTD profiles between surface and 800m depth during field tests in 2012 and 2013. The satellite transmission worked perfectly during the second field test (still ongoing) for the two AITPs deployed in the vicinity of the North Pole in April 2013. The surface unit processor (persistor) proved to be quite reliable.
Improvements are needed for expanding the longevity of the system by increasing battery supply and/or reducing the energy consumption. In the actual situation the autonomy of the ocean profiler is limited to 3 months and should be increased by a factor of 2 at least.
The area selected for these two preliminary field tests located between the North Pole and Fram Strait is ideal for launching the first ACOBAR AITP cluster providing long-term navigation capabilities for a sea glider operating under sea-ice. The ice divergence is not creating any serious problem in distorting the AITP cluster in any significant ways.
From a scientific point of view, there is a great interest in exploring the transition zone between the North Pole and Fram Strait which is an important part of the Arctic transpolar sea-ice drift. A sea glider would provide the missing information concerning the surface layer enriched with brines at freezing point and 34psu salinity in some regions that are actually difficult to map. Also a sea glider would provide missing information about the surface layer still fresh in the vicinity of the North Pole and more salty and colder further south heading in the direction of Fram Strait. It seems like there is a frontal region located at about 87°N where salinity is increasing drastically. All CTD profiles provided by AITP ocean profilers, indicated the presence of a well-defined Atlantic layer at about 300m depth. Based on AITP CTD profiles, there was no sign of a double diffusion mechanism occurring above the Atlantic core and within the main thermocline as it was identified in the central Arctic Basin recently. But this might be due to lack of vertical resolution of the AITP CTD profiles. A sea glider could overcome this difficulty.
It was also interesting to observe significant divergent sea-ice motions related to atmospheric forcing with low pressure systems tending to create convergence and high pressure systems creating some divergence in sea-ice dynamics. But on the long term these converging-diverging sea-ice motions did not alter significantly the geometry of the AITP cluster. In addition to the AITP, we deployed IMB (Ice Mass Balance) systems to provide information regarding temperature profiles across the sea-ice and sea-ice thickness evolution as a function of time. But this is not part of this report. However as an integral part of AITP it was interesting and useful to collect basic meteorological data such as air pressures and air temperatures versus time at surface (the meteorological sensors are located within the AITP surface unit). Both measurements proved to be reliable and could be used to complete networking information such as IABP (International Arctic Buoy Project) and to serve for calibrating satellite sensors. For our application it was interesting to document the highly correlated sea-ice diverging motion and surface air pressures transiting from low to high pressures during a single event. It was reassuring and important to observe that phenomenon had no significant impact on the AITP cluster geometry as far as navigating a sea glider under sea-ice with an AITP cluster is concerned.
The next phase would consist in deploying a full AITP cluster in the vicinity of the North Pole in springtime and to let it drift southwards in direction of Fram Strait together with an acoustic sea glider operating inside the AITP cluster. The AITP would provide all the necessary information to the sea glider operating under sea-ice constantly and releasing data to any of the AITP as soon as it would get within the acoustic range for communicating with the acoustic modems. The AITP grid could be of the order of 100km. The experiment would last 6 months and all the equipments could be retrieved in Fram Strait the following September.
Results from WP3: Fram Strait Ocean Observatory
The acoustic experiments
The Fram Strait acoustic network for tomography, navigation and passive acoustic were deployed in 2010. The system was recovered in 2012. Data, including acoustic recordings and engineering data, from the recovered instruments have been scanned and processed. The recovered instruments in A, B and D provide full year acoustic receptions from 16 hydrophones and engineering data required for clock correction and mooring motion correction. In total data from five hydrophone arrays each consisting of 4 hydrophones provided acoustic data. Except for the first deployment in D (2010-2011) all the arrays had a 9.6 meter spacing between their hydrophones. This spacing allowed for beam forming.
After the engineering data was processed, the sound recordings are corrected for mooring motion and clock drift. The recordings are then analyzed with the aim of identifying ray arrivals of the frequency sweep signals emitted by the sound sources. Matched filtering is used to compress the 1-minute sweep signal, thus optimizing the signal to noise ratio. Beam forming provides information about the arrival angle of the wave front. The estimator-correlator accounts for the incoherence of recorded sweep by using sensible estimates of the coherency bandwidths of the received signal in time and frequency. The data processing in ACOBAR is a clear improvement from the DAMOCLES. The improvement is partly due to the experiment design, which had a spacing of 96 m.
The processing described above was carried out for the ACOBAR data and the time series of travel times and arrival angle along 3 sections has been produced. The first set of travel times was forwarded to WP 4 for inversion to temperature in March 2013. This was later than planned, due to several issues that had to be solved before the data were useful for inversion.
To summarize:
1) The positioning of the transponders and anchor points before the mooring corrections can be made is more complex (more like a puzzle) when the positioning is done from a ship where the interrogating transducer is not fixed to the ship, but hanging over the ship side. Also, the GPS positions given relative to the ship antenna need to be corrected to provide the position of the transducer. Finally, “cleaning of the positioning data” is required to get rid of the data, which are not correct. This is tedious and time consuming work.
2) The most serious problem was that the source functions describing the signal transmitted from AWI and NERSC 1 sources did not follow the specification provided by the “source producer”. When the source signal is not accurately known this leads to problems in the pulse compression and the result is smeared arrival peaks, which results in larger errors in the exact detection of the arrivals. The smeared out arrivals was observed in the DAMOCLES experiment, but it was not known why it was like this. The deviation from the specification was realized after calibration of similar sources owned by Scripps Institution of Oceanography. There is a timing error of up to 50 ms. To get an exact description of the source function the sources should be properly calibrated, but this is expensive. This will be followed up in later projects if funding allows. Otherwise a synthetic description of the signal will be formulated using measurements in the Laboratory.
3) The oceanographic profiles in the open ocean part of the Fram Strait makes the arrivals come almost at the same time like a cluster of un-resolvable arrivals. A significant effort was carried out to optimize the signal processing.
The C mooring is not yet recovered and dredging for the acoustic source in mooring C will be performed as part of the Norwegian funded UNDER-ICE in 2014. All transponders around C mooring were recovered.
Use of gliders and oceanographic moorings
A total of eight glider missions were carried out during the ACOBAR period, both in summer and autumn seasons. During each mission the measured data as well as engineering data of the mission were transmitted in the near-real time to the base station operated by Optimare in Bremerhaven. Gliders were deployed typically for 2-3 months with turn around the vehicles in September, usually during KV Svalbard autumn cruises. All summer deployments were successful with gliders recovered after the end of a mission while during autumn deployments three gliders were lost after between 1 and 2 months due to technical problems (broken antenna or battery failure). However, also prematurely finished autumn missions delivered significant number of hydrographic profiles and RAFOS receptions. Gliders were operating in the eastern and central Fram Strait, mostly in open waters and at the ice edge.
Developing operational capability for the near-real time (NRT) data transfer from the moored array was focused on two main aims: to test the long-range acoustic data transfer between moorings and to develop the communication mooring with the surface unit capable of satellite data transmission. To achieve the first goal, three low-frequency ling-range acoustic modems, the HAM.nodes were tested in Fram Strait in different configurations. Since acoustic data transmission over a typical range between moorings of O(30 km) proved to be unreliable, the distance between long-range modems was reduced by adding a relay-link mooring with additional modem in a half-way between instrumented moorings.
ACOBAR deployments of acoustic modems revealed severe difficulties in achieving a long range communication between the Fram Strait moorings over a whole deployment season. Operation time of the units in many cases was limited due to system failures and thus the tests were done over shorter periods. Moorings equipped with acoustic modems were located in highly energetic area with strong currents, in consequence moorings were rarely in vertical positions due to strong diving. Environmental conditions and a depth of sound channel varied significantly between seasons and signal to noise ratio was relatively low in most cases. The modem technology was improved in the meantime but several options were still lacking during the ACOBAR period. The large number of packets per telegram led to the loss of several datasets during sending attempts during the first long-term deployment. The ACOBAR modem deployments can be concluded the horizontal long-range communication might be possible over a range of about 10 km in Fram Strait if modems could be held in one depth with only little variation. Due to strong variability of environmental conditions, handling of transmission schedule should be improved by checking sound channel properties and depth information first before transmissions. The model technology is promising but still not mature enough for data transmission over the ranges longer than 10 km.
The NRT data transfer from moorings to the land station was addressed by a design of the field experiment with two different systems. The first system consisted of the underwater winch which was equipped with the two different profiling CTD tops: one built on a base of the NEMO float and one being the generic NGK profiler. The stand-alone mooring with a tethered standard NEMO float was also tested but it was found that it could not negotiate the strong currents therefore was not able to reach the surface. The profiling winches used for test deployments of several weeks worked reliably according to the programmed schedule. However, the first tests of the NEMO-based CTD profiler in 2010-2012 were not successful. The profiler either got lost or was damaged (flooded or broken) during three subsequent deployments and no communication was established with the land station. Only during the last deployment in 2012, the heavy instrumented NGK profiler provided 4 profiles per day (213 in total) and reached the surface during nearly each ascension. Data were also partially sent to the land station. Therefore, based on ACOBAR output, this technology will be used for future development of NRT data transfer from the Fram Strait moored observatory.
The NRT data transfer was also achieved during the glider missions in Fram Strait in 2010-2012. During all missions, the full engineering and scientific data sets from the upper 1000m layer were transmitted by the glider during surfacing after completion of each single dive. The temperature, salinity, dissolved oxygen and fluorescence profiles and sections were processed and visualized in the near-real time during the on-going glider mission. There is on-going development focused on application of a short-range acoustic modem dedicated to data transmission between glider and underwater moored instrumentation. Therefore, future solutions for using gliders as data messengers from a moored array are considered as more robust then currently tested moored profiling solutions.
To implement the acoustic positioning and navigation system for gliders and floats for future NRT data transfer from under the ice, three RAFOS sources with frequency 260 Hz were deployed in July 2010, four RAFOS sources in July 2011 and seven in July 2012. Additionally three tomographic sources, deployed in 2010 (and partially redeployed in 2011) in a triangle covering the northern Fram Strait were programmed to provide the RAFOS transmission between tomographic sweeps. During all missions, gliders collected RAFOS receptions and calculated navigational solution based on RAFOS signal using the built-in RAFOS hardware and the dedicated firmware developed by the APL-UW group led by Craig Lee. In summer 2010 the glider received 447 RAFOS transmission with maximum ranges for RAFOS source located in the ice-covered area up to 250 km. During the summer mission in 2011 the glider collected altogether 298 valid RAFOS receptions, used to calculate 68 positions, of which 70% with error less than 10km and 25% with error less than 5 km. The accuracy of positioning improved for gliders equipped with a new type of RAFOS hardware when during the autumn mission in 2011 72 % of 99 reliable positions had an error less than 10 km and 40% less than 5 km. 147 valid positions were based on 483 acoustic receptions collected by the glider in summer 2012 (40% with error < 10km). The accuracy of acoustic navigation is acceptable for under ice mission but can be still improved by using more accurate clocks in sound sources and better synchronisation between sources and glider RAFOS hardware.
Results from WP4: Assimilation of tomography data
The aim of the work in WP4 is to invert acoustic travel time data to temperature fields and develop of assimilation schemes for use of acoustic travel time data in ocean models. The acoustic measurements are different from other oceanographic measurements by its integrative characteristics. The challenges of the assimilation schemes are to understand how the wide arrival pulse reflect, how to quantify the distance between the observed broad pulse with the ray predicted arrivals, and how can we cope with this inversion and assimilation.
Acoustic travel time measurements are inverted to range averaged sound speed profile which again is averaged over depth to a range-depth averaged sound speed to achieve a very high accuracy. A simple scale factor of 4.5 m/s/OC can be used for conversion of sound speed variations to temperature variances in the Fram Strait. The accuracy of the conversion is within a few percentages. In the case of assimilation the temperature and salinity will be updated within its dynamical constraints.
A numerical investigation show that the shape of the single, broad arrival pulse of around 100 ms width arises from a combination of the minimal dispersal of the acoustic arrival pattern cased by the fairly weak sound speed channel in the Fram Strait and internal waves causing scattering of the acoustic field. The internal wave scattering results in proliferation of eigenrays that essentially fill the water column between the source and receiver – this indicate that the individual ray paths no longer have a meaning. This points towards the use of Time Sensitive Kernel as described by Skarsoullis et al. 2009 which has not yet been used for inversions. This methodology can also create machinery for future assimilation. This approach will be addressed in upcoming project proposals.
The inversion scheme developed is based on vertical EOFs established based on the high-resolution ice ocean model developed within the DAMOCLES. The horizontal variability is handled by sinus and cosines. Most important, the principle of identification of specific arrivals with specific rays were abandoned the. The strategy was to quantify the oceanic variability apparent in acoustic arrival patterns consisting of wide acoustic pulses of some 100 ms in width. Since all computed rays arrive at the same time, it does not matter which ray is assigned to which particular peak, so first order. Only the peaks from the single-wide pulse were employed for this inverse. Large 100 ms uncertainties were assumed for the data, however, to account for the lack of ray identifications. The inverse problem is saved by the essential resolution of the integrating measurements, however. Furthermore, the large amount of data minimizes the impact of the large data uncertainty (the data being picked-peak travel times).
Ultimately, the formal uncertainties in temperature estimated from the inverse were about 50m°C. The high-frequency variability of the temperature time series was larger, with a 78m°C RMS. The corrections to the Fram Strait model in sound speed were then used to estimate absolute temperature, using a conversion factor of 4.5 m/s/°C. The estimated temperature variations were in remarkable agreement with earlier estimates by Skarsoulis et al. 2010, who used a completely different inverse approach. In future experiment it may be that employing a deep receiver at 1700 m may result in a stable resolved ray which may double the available information for the inversion.
The inversion method is used in a numerical study to compare the use of point measurements and tomographich measurements either separately or combined. It is shown that for obtaining an estimate of an average of a scalar quantity along a section across Fram Strait, optimal measurements are a combination of point and line-integral data types. This combination reduced the uncertainty of the estimate by about a factor of three. An equivalent reduction of uncertainty using only point measurements would require a roughly order of magnitude increase in the number of point measurements. This result demonstrates the fundamental motivation behind employing acoustic tomography in conjunction with other data types.
Our hypothesis of ACOBAR is that the tomography data will improve or constrain the model estimates for temperature within Fram Strait, hence lead to a more accurate picture of the exchange of heat between the Atlantic and Arctic Oceans. As can be seen in the sections below, considerable progress has been made toward this goal.
The work to implement data assimilation with line-integral constraints (tomography) is technically challenging. Starting point of the work in ACOBAR is based on the results from the single track experiment conducted in DAMOCLES from 2008 to 2009.. The oceanographic conditions in the Fram Strait cause the acoustic arrivals to come in a cluster. This makes it impossible to use methods based on separation of individual arrivals. This influences how to establish the assimilation and inversions schemes. In DAMOCLES this initiated an analysis of how to establish the measurement matrix and how to compare the modeled arrivals with the observations. The approach of identifying and comparing individual rays were abandoned in both the assimilation and the inversion approach (Skarsoullis, et al 2010). In the assimilation another problem was encountered. The modeled observation matrix (predicted arrivals) did not match the acoustic observations either in the size of variability or the seasonal variability (Haugen et al. 2010, Sagen and Sandven, 2006, Bertino et al. 2009, Sagen et al. 2010 a,b, Haugen et al. 2010). The ice-ocean model system developed within DAMOCLES for the Fram Strait, and status in the use of Ensemble Kalman filtering with acoustic measurements is provided by Sagen et al. (2010 a,b).
The major findings in the experiments was that
1) Acoustic arrivals in the West Spitzbergen Current come in as a broad pulse of un-resolvable arrivals making it impossible to detect arrivals and identify a ray arrival to it.
2) A “brute force” inversion based ray arrivals was developed and provided depth-range mean temperatures
3) Modeled observation matrix were discussed and compared to the observations and used as a tool to validate the high resolution ice-ocean model and the Topaz 3 model.
4) Distance between modeled and observed acoustic travel times were difficult to quantify due to point 1) and also because the ice-ocean models are not representing the ocean good enough and the modeled observation matrix is strongly dependent on the interpolation routines in both the vertical and the horizontal.
The activities in ACOBAR has first of all aimed towards providing new, and essential ground work for data assimilation. In this first phase of WP 4 the travel time data originated from the 2008-2009 experiment has been used (Haugen et al. 2010). This was because acoustic data was only available after the cruise in 2012. In the work described in this report the nature of the unique acoustic properties within Fram Strait has been further elucidated, preliminary inversions of the acoustic data to obtain estimates of temperature have been completed, and the expected impact of the acoustic data compared to such data as the set of Moored Array thermistors has been quantified. We have a much better understanding of the expected impact of these data on the numerical model, hence how to test the integrity of the eventual data assimilation results.
The second issue has been that it has been difficult to precisely quantify the differences between ocean model predictions and observations such that these differences can be used to correct, or constrain, the ocean model. The discussion and resolution of this latter issue comprises a large part of this report.
Results from WP5: Dissemination and exploitation
The ACOBAR project website (http://acobar.nesc.no) has been established and populated with information about activities in the project. The other participants have contributed with various dissemination material for the website. AWI has established a dedicated website for glider activities in Fram Strait (http://www.awi.de/?id=6021). It contains background information on gliders, technical description of glider types used in ACOBAR (with a special focus on Seagliders used in Fram Strait), a summary of state-of-art regarding the acoustically navigated gliders and a description of the planned multipurpose acoustic system for glider and float navigation in Fram Strait. The picture gallery and a collection of relevant publications and links related to different glider activities are also provided.
The first version of the dissemination and use plan was provided in period 1 (D5.2). An updated version of this plan has been prepared at the end of the project.
Dissemination activities include publication of peer review papers in science journals, conference proceeding, poster presentations, oral presentations and meetings with users. A detailed list is presented in Table A2.
Links to other projects developing underwater observing systems. ACOBAR has been presented to the ESONET Network of Excellence. ACOBAR will contribute to ESONET by an Arctic demonstration project in 2010-2011. ACOBAR has also established links to Norwegian Defense Research Establishment, Kongsberg Maritime and other companies working with AUVs and underwater acoustic systems
Results from WP6: Transfer of technology and knowledge
Activities at NERSC
The transfer of technology and knowledge in ocean acoustics has been obtained through close collaboration with the partners Scripps Institution of Oceanography (SIO) and Woods Hole Oceanographic Institution (WHOI) in USA. The activities have included exchange visits between Norway and USA, training in use of acoustical systems, planning and implementation of field experiments, data processing, presentation at workshops and conferences, and publication of papers. Students and young scientists have been involved. The activities started in the DAMOCLES project (2005-2009) and has continued as part of ACOBAR from 2008.
In April 2009 Hanne Sagen and Stein Sandven from NERSC visited WHOI and Webb Research Coorporation as part of the preparation of the acoustic tomography system. The visit was combined with an ACOBAR progress meeting (14 April) and a Workshop (15 April). In August 2009 Dr. Peter Worcester from SIO visited NERSC in Bergen and Longyearbyen and participated in the cruise with RV Håkon Mosby. The tomography moorings were recovered and one year of tomography data were obtained. Dr. Peter Worcester gave a guest lecture on ocean acoustics when he was in Bergen.
Hanne Sagen and Svein Arild Haugen from NERSC visited Scripps Institute of Oceanography (SIO), USA for 14 days (22 October - 6 November 2009). The aim of the visit was to process and start analyzing the acoustic data from the first Fram Strait experiment 2008-2009. Data analysis was completed after returning to Norway. Svein Arild Haugen used this work in his Master thesis completed in April 2010.
• The cooperation with WHOI and Scripps has resulted in several publications and reports
• Advanced acoustic receiver and source technology has been purchased from Scripps Institution of Oceanography, and Teledyne Webb Research Cooperation, respectively. NERSC personnel have been trained in deployment, maintenance, testing and deployment procedures by engineers and scientists at the SIO. SIO have participated in the field operations and they have worked closely with scientists form NERSC, University in Bergen (UIB), and AWI.
• The most efficient transfer of knowledge takes place during planning and execution of field experiments. Peter Worcester and Andrey Morozow, WHOI/Webb Research participated in the RV Håkon Mosby first deployment cruise in 2010. Lora Van Uffelen (Post Doc) from SIO and Peter Koski (Engineer) from WHOI (communication) took part in the KV Svalbard Cruise (2010). Peter Worcester and Lloyd Green participated in the KV Svalbard cruise, 2011 (recovery/redeployment). Aquatec Ldt (2009, 2011) and NAXYS (Norwegian SME) (2008, 2009, 2010) have participated in several cruises. ENSTA and NKE participated this year in Håkon Mosby cruise.
• Brian Dushaw, Applied Physics Laboratories, Seattle, is visiting scientist, funded by Fullbright fellowship and ACOBAR, from 1 September till end of February. Dushaw has long experience in inversion of acoustic, and he is working together with Hanne Sagen on assimilation/model oriented inversion in WP 4. Dushaw took part in the field experiment in 2011 and 2012.
• Scientific visits from SIO and WHOI to NERSC: Dr. Peter Worcester and Mathew Dzieciuch
Activities at AWI
Regarding use of gliders in Arctic waters. AWI has cooperated with Applied Physics Laboratory, University of Washington, USA (APL-UW). AWI uses the Seaglider produced by APL-UW, and the first successful testing of this glider in the Fram Strait was done in 2008. An adaptation of the Seaglider for under-ice operations has been done under a collaboration with the Integrative Observational Platforms Group (APL-UW) led by Craig Lee. The first winter long Seaglider deployment in Davis Strait from September 2008 to February 2009, performed under the US National Science Foundation project 'An Innovative Observational Network for Critical Arctic Gateways'http://iop.apl.washington.edu/projects/ds/html/program.html proved feasibility of extended under-ice mission where Seagliders navigate using acoustic signals received from moored RAFOS sound sources. In ACOBAR, both conventional RAFOS sound sources and tomographic sources will provide RAFOS navigation signals and the RAFOS technology provided by the APL-UW will be used for under-ice navigation of Seagliders in Fram Strait. OPTIMARE acts as the link between Seaglider manufacturers and the end users (scientists) to promote the exchange of information and requirements and to assist the field operations.
An adaptation of the under-ice Seaglider and acoustic navigation technology used in Davis Strait for Fram Strait has been done under a collaboration with the Integrative Observational Platforms Group (APL-UW) led by Craig Lee. The first winter long Seaglider deployment in Davis Strait from September 2008 to February 2009, performed under the US National Science Foundation project 'An Innovative Observational Network for Critical Arctic Gateways'http://iop.apl.washington.edu/projects/ds/html/program.html proved feasibility of extended under-ice mission where Seagliders navigate using acoustic signals received from moored RAFOS sound sources. In ACOBAR, both conventional RAFOS sound sources and tomographic sources will provide RAFOS navigation signals and the RAFOS technology provided by the APL-UW was used for acoustic positioning and navigation of Seagliders in Fram Strait. In collaboration with the APL-UW group the system used in Davis Strait was adapted for Fram Strait, taking into account the environmental conditions and different frequency of available RAFOS sources.
Activities at UPMC, ENK, ENSIETA and AQUATEC
There has been extensive transfer of technologies between UPMC, ENK, Aquatec and ENSIETA on the development and testing of the AITPOPS cluster with glider. ENSIETA has been recently exchanging data and knowledge with AWI, this cooperation helps ENSIETA to design its long range positioning algorithms using AWI's RAFOS data acquired in the Fram Strait
Activities at ACSA
The expertise of ACSA in development of underwater navigation system has been used in analysis of the glider navigation data obtained during the experiments in the Fram Strait
Potential Impact:
ACOBAR has developed and demonstrated sensors and platforms that can improve the in situ observing network considerably compared to the present situation. The ocean observing system provided by ACOBAR is expected to have impact on a research and monitoring of the polar regions by improving the in situ observations of physical and biogeochemical parameters of the marine environment. Several thematic areas will benefit from ACOBAR such as ocean and climate monitoring, marine transportation and safety of operations, ocean acoustics including navigation and communication, oil and gas exploration, marine resource management, and protection of marine ecosystems in the Arctic. Development and implementation of ocean observing system for this wide range of user groups requires active promotion and dissemination of the results from ACOBAR after the project is completed. The need to enhance the in‐situ marine monitoring capacity by on global scale has been recognized by GOOS, GCOS and others, while on European scale EuroGOOS is the main coordinating body for operational ocean monitoring and forecasting. All the 34 EuroGOOS members from 16 countries play key roles in improvements of the in‐situ marine monitoring in the European seas including the Arctic. One condition for improvements of the observing systems is, among others, new investments in RTD activities that would deliver new, more efficient, robust and cost effective sensors to equip different platforms such as buoys, gliders and floats. The ACOBAR project has therefore demonstrated a significant improvement of the ocean observing systems in the Arctic seas. The next step after demonstrating the capacity of the ACOBAR system is to implement a larger number of platforms that can operate over large parts of the Arctic Ocean. This requires cooperation with USA, Canada, Russia and other nations to install and maintain a network of platforms in the Arctic. It also requires that research and monitoring institutions in these countries increase their funding of long-term Arctic programmes that can sustain the monitoring system.
Multinational efforts are required to build an ocean observing system in the Arctic based on a network of in situ sensors and platforms. It is necessary that European institutions cooperate with USA, Canada, Russia and other countries in order to implement and sustain a pan-Arctic ocean observing system.
Increased availability of standardised in-situ data
ACOBAR has provided oceanographical data from the AITPs, the gliders and the other platforms which are interoperable with other data repositories for oceanographic data. The data collected are delivered in standard formats established by the oceanographic community. These formats are used in operational observing systems such as the ARGO programme (http://www.argo.ucsd.edu/) and are adopted in many European projects (SeaDataNet, EMODNet, MyOcean, GISC). SeaDataNet is coordinating European ocean and marine data management, and defines standards and software for implementation of this work (http://www.seadatanet.org/Standards-Software).
Data from Ice-tethered profilers (ITPs) have been established in the Arctic by WHOI almost 10 years ago, and data from these profilers are provided in standard formats (http://www.whoi.edu/page.do?pid=20781). The AITPs in ACOBAR provide data in the same formats. At present data from ITPs and gliders are provided as profiles to the GMES Marine Core Service operated by MyOcean & MyOcean2 FP7 projects using some agreed processing methods developed within FP5-MFSTEP, FP6-MERSEA and FP7_MyOcean projects. In the GROOM project, a data management plan is under development where a single access point for all European gliders will be established. The glider data from ACOBAR will be included in this plan.
Glider data from ACOBAR will also be linked to the OceanSITES program, which is the global network of open-ocean sustained time series sites, called ocean reference stations, being implemented by an international partnership of researchers (http://www.oceansites.org.). OceanSITES provides fixed-point time series of various physical, biogeochemical, and atmospheric variables at different locations around the globe, from the atmosphere and sea surface to the seafloor. The program’s objective is to build and maintain a multidisciplinary global network for a broad range of research and operational applications including climate, carbon, and ecosystem variability and forecasting and ocean state validation.
Establishment of web-interface to marine data repositories is now standard services provided by many data producing institutions. In Europe several ongoing project are developing web-based systems for marine data. NERSC is coordinating the NETMAR project (Open Service Network for Marine Environmental Data), which develops tools for searching, downloading and integrating satellite, in situ and model data from ocean and coastal areas. It is a user-configurable system offering flexible service discovery, access and chaining facilities using OGC, OPeNDAP and W3C standards. Data from ACOBAR will be provided in web-system that is fully compatible with other European projects (i.e. NETMAR, SeaDataNet, EMODNET, etc.).
Reduced cost of data collection system in support of fisheries management;
There is a great need to develop and improve the temporal and spatial scale of in situ oceanographic observations, and this requires more cost-efficient observing systems to be operated from many different platforms (ships, floats, gliders, ice buoys, etc.). One of the goals of ACOBAR was to develop and demonstrate more cost-effective systems for use in the Arctic. CTD instruments (Conductivity, Temperature and Depth/pressure) are fundamental for observation of physical, biological and biogeochemical ocean parameters. They are measured in their own right to understand ocean processes such as water mass movement, but are also required to give physical context to a huge variety of other measurements. Acoustic methods require sound speed profiles obtained by CTD data. As these parameters are so crucial, they are likely to feature extensively in the MSFD monitoring requirements. From an operational perspective, these basic observations are required as standard across a broad range of applications from climate to fisheries, shipping, and energy exploitation.
CTDs and other sensor packages must be operated from platforms that can improve the temporal and spatial sampling of data. In ACOBAR, this was demonstrated by using ice platforms in combination with glider, which can be operated autonomously over long time. This will be the most cost efficient method for data collection in the Arctic Ocean. In the MIZ and open water areas, which are most important for marine resources and fishery management, gliders can in principle be operated year round with data transmission when the gliders are at the surface in open water. Continuous monitoring of physical, biological and biogeochemical will provide more data to support fisheries management. Ship surveys will still be needed at certain times of the year for water sample collection and other measurements that cannot be done with automated sensors. But, gliders and ice buoys with profiling instruments can take over much of the regular data collection from sensor that can operate automatically over long time. A European infrastructure project (GROOM) has been established to develop use of underwater gliders for collecting oceanographic data (http://www.groom-fp7.eu/doku.php). This new infrastructure shall be beneficial for a large number of marine activities and societal applications, which can be related to climate change, marine ecosystems, resources, or security and which rely on academic oceanographic research and/or operational oceanography systems. Several partners in ACOBAR are also members of GROOM (UPMC, AWI, NERSC).
Advance competitiveness for European Industry's & particularly SME's within the Marine sensing sector;
Companies representing offshore industry, underwater technology and manufacturer of observing systems have been involved in ACOBAR both as partners, subcontractors and users. These companies will be the key drivers for developing observing systems to be used in the Arctic in the future. As an example, Scripps Institution of Oceanography will provide expertise and technology in underwater acoustics. Furthermore, Teledyne Webb Reserch, which is a world leading manufacturer of gliders and sensors for use on gliders, will contribute to build up European expertise in this technology. Users of the observing systems will provide useful feedback to the companies based on the testing of the systems in the field trials. The experience gained from ACOBAR will strengthen the competitiveness of the European companies in the marine observing sector. Experience from the Arctic field trials will be a competitive advantage for companies planning to be in the forefront of Arctic underwater technology.
Enable better cooperation between key sectors (Manufacturing Industry, ICT, Maritime Industry, Marine Science, Fisheries etc.);
ACOBAR results will continue to be disseminated to representatives from the manufacturing industry (Teledyne Webb Research), maritime industry / shipping working with focus on the Arctic (Norwegian Coastal Directorate, Centre for High North Logistics), marine science (representative from EuroGOOS), fisheries (Norwegian Fisheries Directorate) and ICT (SeaDataNet, NETMAR). The technologies and results from ACOBAR are of interest for all theses sectors because of the increased capacity to collect in situ data. It is envisaged that better cooperation between the sectors will emerge since they will have benefit from the same technologies and their data producing capabilites.
Data dissemination from ACOBAR will be interoperable with other European initiatives to coordinate and sustain collection, standardization, archiving and dissemination of marine data, in particular SeaDataNet, EMODNET and GISC. Data from ACOBAR will for example be used by the Arctic Marine Forecasting Centre, which is the Arctic Node of the Marine Core services provided by MyOcean (http://myocean.met.no/). ACOBAR will contribute to build up observation technology and improve data collection which is relevant for all these sectors.
Support implementation of European Maritime Policies (MSFD, CFP, IMP, etc.);
The European Maritime policies aims to provide the mechanisms for clean and healthy marine environment, and ensure that exploitation of marine resources is done in a sustainable way. This is particularly important in the Arctic where exploitation of energy and other resources are expected to grow in the future. The policies are also focused on adaptation to climate change, which has significant impact on environment and ecosystems. This is also particularly important in the Arctic Ocean where global warming is most pronounced and the impact on environment is significant.
The Marine Strategy Framework Directive (MSFD) includes an initiative for the development and actual implementation of an overarching European Marine Observation and Data Network (EMODNet). EMODNet is considered as a network of existing and developing European observation systems, linked by a data management structure covering all European coastal waters, shelf seas and surrounding ocean basins. EMODNet is coordinated at EU level by a task team lead by EU DG MARE and tuned with other European directives (MSFD, INSPIRE) and large-scale framework programmes on European and global scales (GMES and GEOSS), that urge access to, and exchange of, environmental data and information. New regulations from MSFD require observations to determine good environmental status. In addition, the Common Fisheries Policy and regulations regarding noise also need data. Therefore ACOBAR results regarding use passive acoustic sensors to monitor the ambient noise and how it varies with space and time. There are operational demands as well, with fisheries, renewable energy, oil and gas, weather and ice forecasting, shipping and defence all requiring in situ observations to support daily operations as well as long-term planning. Large-scale observations will increase understanding of ocean processes, biology, biogeochemistry, climate change and marine health, as well as improving validation of models. The success of the ARGO programme has shown that large scale in situ observations are possible and hugely beneficial..
Safety of marine operations is one of the pillars of European Maritime Policies, and thus a major reason for building up the GMES marine services (http://www.myocean.eu/web/72-marine-safety-description.php). These services provide large-scale ocean and sea ice monitoring from satellite data and short-term ice-ocean forecasting for the whole Arctic with about 20 km grid cells. The forecasting products are delivered by the TOPAZ system at NERSC (http://topaz.nersc.no) and operated by Norwegian Meteorological Institute through the Arctic Marine Forecasting Center (http://myocean.met.no/). The ocean modelling and forecasting systems require improved in situ observing network in order to validate the models and assimilate data into the models. Data from ACOBAR will therefore be a valuable contribution to the Arctic Marine Forecasting Center, in particular the improved access to near real time data
ACOBAR has demonstrated technologies for in situ data collection which is required for development of ocean and sea ice monitoring and forecasting services. These services are needed to ensure safe and economic marine operations in polar regions where sea ice and icebergs represent a significant risk factor for all types of operations.
Assumptions and external factors affecting the impact
The overarching assumption is that there will be a growing demand for in situ ocean data from the Arctic Ocean. There are a number of reasons for this:
(1) There is a need for more physical oceanographic data in the Arctic because the ocean and sea ice plays an important role in the climate system, and this role is not well understood. The ocean and sea ice models, including climate model, need more data to improve the models and validate the model results.
(2) In addition to physical data there is also need for more biochemical data to understand how the ecosystems work, how the marine resources develope under a changing climate, and how human activities affect the ocean environment.
(3) Growing industrial and shipping activities will have demand for monitoring and forecasting systems, which will require observational data
(4) Satellite earth observation, especially the GMES programme, requires more in situ data in the polar oceans for validation of satellite retrieved data and provide subsurface and deep sea data that cannot be obtained from satellites.
(5) Satellite communication will have impact on how much data can be transmitted from various observing systems to the users. It is assumed that such services will improve and grow fast, offering more and more capacity to transfer data
(6) The growth in industrial activities in the Arctic will require development of underwater technology, and companies working with such technologies are the backbone for building new in situ obsvering systems. The strength of this industry is an important factor to push development an dimplementtaion of new obervsing systems
(7) Arctic marine transportation is expected to grow in the future as a result of reduced sea ice. This transportation is driven by the global market and the comparative advantage of the Arctic transporation routes compared to other routes.
(8) The demand for oil and gas exploitation from Arctic regions will be a main driver for future activitites and will define among others the extent to which Arctic industry, shipping and underwater technology will be developed.
(9) Research and monitoring activities undertaken by various counties working in the Arctic will have impact on the demand for the observing systems developed in ACOBAR.
(10) Geopolitical situation and legal framework will also have impact on the interest of different countries and industries to be present and make investments in the Arctic.
List of Websites:
Public website: http://acobar.nersc.no
Contact details:
Prof. Stein Sandven
Director
Nansen Environmental and Remote Sensing Center (NERSC)
Tel: +47 55 20 58 00
Fax: +47 55 20 58 01
E-mail: stein.sandven@nersc.no
The overall objective of ACOBAR has been to develop a system for environmental monitoring of the interior of the Arctic Ocean by use of acoustical methods, in situ point measurements and numerical models including data assimilation. During annual field experiments fro 2008 to 2012, ACOBAR has demonstrated several acoustic methods, including tomography, data transmission and communication to/from underwater platforms, and navigation of gliders. The ACOBAR results provide complementary methods to the ARGO system, which cannot be used in ice-covered seas, based on platforms located under the sea ice. Data collection and transmission from the water column and the seafloor has been tested using acoustics methods on moorings, gliders and ice-tethered platforms. ACOBAR has developed methods that can be used in the polar regions, contributing to filling gaps in the global ocean observing system. ACOBAR has conducted field experiments with acoustic sources and receivers, as well as gliders in the Fram Strait and drifting ice-tethered platforms in the high Arctic Ocean. An array of acoustic sources and receivers was operated over two years for measurement of acoustic travel time, and methods have been developed to retrieve integrated ocean temperature from the travel time data. Acoustic tomography in combination with in situ data and numerical models is an innovative method to obtain integrated 3-D fields of temperature, transports and heat fluxes in the ocean. Long-range acoustic signal propagation has been used to develop localisation methods for gliders when they are in sub-surface position. These methods will be the basis for acoustic navigation of gliders, which will be required for gliders to operate under the ice-cover. Also data transmission from fixed moorings via acoustic modems to the surface for downloading from ships or for satellite transmission has been tested and useful experience has been gained for further development of underwater data transmission. The acoustic technologies aim to be used for transmission of multidisciplinary data from underwater observatories, which are presently under development in many sea areas. Transfer of technology and know-how between USA and Europe and between industry and research institutes has been an important component of the project. The exchange of personnel between the partners, workshops and meetings between scientists, engineers and students have made it possible to build up expertise for planning and implementation of complex field experiments and obtain unique data sets. The first results of the project have been published in referee journals and popular science articles as well as presentations at conferences and workshops. Further publications and other dissemination of results will be done after the project is completed.
Project Context and Objectives:
The Arctic Ocean is becoming increasing important as a result of the global warming and the growing economical and political interest in the region. Sea ice reduction is facilitating resource exploration, marine transport and other economic activities in the Arctic and sub-Arctic regions. It is anticipated that over the coming decades the Arctic is likely to attract substantial investment, potentially reaching $100bn or more. It is therefore important to improve our understanding about the climatic and environmental changes in the region, along with their socioeconomic impact. On this background the Arctic is an area of growing strategic importance for Europe, as shown in a number of policy documents from 2008 to present (http:// http://eeas.europa.eu/arctic_region/). EU’s Arctic policy is focussed on three policy areas: (1) protecting and preserving the Arctic in unison with its population; (2) promoting the sustainable use of resources; and (3) international cooperation.
The Arctic Ocean is poorly observed and monitored compared to the other world oceans. It is clear that the issues associated with Arctic marine change transcend national boundaries and science disciplines. No one nation has the sovereignty, expertise, knowledge, logistical capability or budget to individually tackle these challenges head on; a truly international and integrated scientific effort is needed. Understand Arctic marine change needs a technological, multidisciplinary, and integrated approach. While satellite observations are important for monitoring the ocean surface including sea ice, there is severe lack of in situ and subsurface data for the Arctic Ocean. Such information is essential for research on climate, marine resources, shipping (tourist and commercial), oil and gas exploration, support to operations and protection of the environment. It is therefore of high priority to create and sustain an Arctic Ocean observing system that can provide services to different users groups such as policy makers, climate researchers, marine scientists, environmental agencies, management of natural resources, shipping and other maritime activities. At present collection of in situ ocean data in the Arctic seas is generally performed by icebreakers, aircraft and submarine expeditions, which are all very expensive and not suitable for providing regular, long term monitoring data. A buoy programme with Ice-Tethered Profilers has started, but covers only a limited part of the Arctic Ocean (http://www.whoi.edu/page.do?pid=20756).
In recent decades, the Arctic Ocean has experienced dramatic changes documented by the record low summer minimum ice extent (http://arctic-roos.org) and the strong reduction of multiyear ice (The Economist June, 2012). Atmospheric warming is a dominant forcing for the melting of ice, but evidence is now building that warmer ocean also contributes to the thinning of Arctic ice (Carmack and Melling, 2011). It is well documented that the ice is getting thinner and more seasonal (Wadhams, 2012), and this reduction in ice coverage will lead to stronger exposure of the surface layer to wind-induced mixing. In the Arctic Ocean a 100-200 m thick surface layer of low salinity and cold water protects the ice cover from more extensive melting by the warmer underlying Atlantic water. Under intensified wind forcing the isolating surface layer can easily destroyed due to mixing with warmer water masses and this process may accelerate the sea ice melt beyond what we have seen so far. The accelerated melting observed during the last years might indicate that this process already is in progress. The heat accumulated in the ocean can also retard the ice growth through the fall and into the winter and this process can be substantially impacted by upper ocean stratification changes (Steele et al, 2010). However, the currently available climate models are yet not able to reproduce the observed accelerated ice melting (Rampal et al. 2011). The weak turbulent mixing in the upper layer and consequent small vertical heat flux allow the existence of a cold halocline in the Arctic Ocean (Fer 2009). However, most of turbulence in the Arctic region is associated with internal waves over deep topography while near-inertial waves, generated by ice pack motions also play an important role for the upper ocean mixing. Recent findings show that increased near-inertial waves generated in the presence of ice pack at the surface will enhance shear and mixing in the Arctic (Rainville and Woodgate 2009). A seasonally ice free Arctic Ocean can evolve towards conditions similar to in the marginal ice zones, and this can lead to a redistribution of water masses, meaning that the isolating halocline layer can be weakened and vertical heat flux from the underlying Atlantic water will increase.
Since the 1990s, in particular during the International Polar Year (2007-2009), ocean observations aimed in monitoring the volume, heat and freshwater fluxes between the Arctic Ocean and the subpolar seas have been significantly enhanced. However, when using available estimates of long-term mean fluxes, the mass and heat budget of the Arctic Ocean still cannot be balanced, mainly because the uncertainties in the Fram Strait fluxes are larger than the total fluxes through the other gateways (Beszczynska-Möller et al., 2011). Fram Strait is the main deep passage through which the ocean mass and heat exchange between the Atlantic and Arctic Ocean takes place. On the eastern side of the strait the northbound West Spitsbergen Current (WSC) transports Atlantic water to the Arctic Ocean, whereas on the western side the East Greenland Current (EGC) transports sea ice and polar water from the Arctic Ocean to the south, towards the Nordic Seas and Atlantic Ocean. The topographic structure of the strait causes a splitting of WSC into at least three branches, of which one recirculates between 78°N-80°N.
An oceanographic moored array has operated since 1997 to monitor the ocean water column properties and oceanic advective fluxes through Fram Strait. An extremely complex circulation pattern has been observed, characterized by a pronounced recirculation and strong mesoscale activity. Since the spatial resolution of the mooring array varies from 10 to 30 km, it is not sufficient for resolving the mesoscale variability and recirculation, and large errors exist in estimates of the oceanic fluxes. At present the moored observing system does not have any surface expression and therefore cannot perform measurements of biogeochemical and hydrographic parameters in the upper water column (0-50 m). Moreover, below 1000m a vertical resolution of measurements is very low which can introduce a significant bias in the vertical stratification and circulation in deep layers. An improved observing system is required, with high temporal and spatial resolutions capable of quantifying the impact of mesoscale currents and determining the influence of the recirculation on net oceanic transports through the strait. Such an integrated observing and model system for long-term environmental monitoring in Fram Strait, combining data from satellites, acoustic systems, moorings, and gliders with high-resolution ice-ocean circulation models through data assimilation has been under development in ACOBAR project (Sagen et al., 2010, Sandven et al. 2012).
The main objective of ACOBAR has been to develop observing systems for environmental monitoring of the deep Arctic Ocean by assimilation of data obtained with acoustical methods including tomography/thermometry, data transmission from underwater platforms, communication and navigation of floats and gliders under the ice-cover. The observing system has been developed and tested in two different areas: the Fram Strait, which is partly ice-covered and partly ice-free, and the high Arctic which is fully ice-covered throughout the year. The Fram Strait observatory consists of moored acoustic sources and receiver for tomography, and gliders combined with ship surveys, moored profilers and satellite observations of sea ice. In the high Arctic, also ice-tethered platforms were used to observe changes of water masses. Improved observation of the heat, mass and freshwater transport in the whole Arctic is important for climate research.
The specific objectives have been:
1) Develop improved 3-D observations of the properties and transport of water masses in Fram Strait using an acoustic tomography array, oceanographic moorings and profiling gliders through data assimilation.
2) Test and assess transmission of long-range acoustic navigation signals and commands to operate autonomous underwater vehicles such as gliders and floats in the Fram Strait.
3) Implement data transmission by acoustic modems from underwater platforms to the surface for downloading to ships or satellite transfer.
4) Exploit the existing array of acoustic sources from ice-tethered platforms for tomographic measurements of water mass properties in the ice-covered Arctic Ocean with real-time data provision via satellite.
5) Dissemination of underwater oceanographic data to users with near real-time capability.
6) Provide the technology to combine the oceanography array to transmit data through Arctic cabled observatories when available.
7) Transfer of technology and know-how between USA and Europe, and between industry and research institutes.
Project Results:
Results from WP1: Design, synthesis and assessment
The acoustic tomography experiment was carried out from 2010 to 2012 in a triangle of transceivers moorings with a moored receiver in the center. This geometry gave a total of six acoustic paths that covered the Fram Strait Marginal Ice Zone. The acoustic transceivers also transmitted RAFOS signals to provide an underwater acoustic navigation system for gliders and floats. Several glider experiments were conducted where the acoustic signals from the sources were used to develop acoustic navigation under ice. This experiment was the first implementation of a multipurpose acoustic network for tomography, passive acoustic monitoring, and navigation of gliders. The primary goal is to improve the accuracy of the heat, mass and freshwater transport estimates through the Fram Strait using a four-dimensional data and model system combining acoustic travel time measurements and ocean data from gliders and moorings with high resolution ice-ocean modelling through data assimilation. Another goal is to observe and explain changes in ambient noise with respect to changes in environmental conditions and increased human activities (Sandven et al, 2012).
The ACOBAR project has successfully shown that network of acoustic transceivers and receivers can be used as a infrastructures for acoustic tomography covering the whole Arctic basin, underwater GPS system for tracking of floats, assistance to navigation of underwater vehicles, and for passive monitoring of marine mammals, human activity and environmental conditions (ice dynamics, wave conditions). This makes the acoustic networks to an important component in a future integrated Arctic observing system (e.g. Sagen et al. 2010, Lee et al. 2010). The interdisciplinary benefit of acoustic components in an integrated Acoustic Observing system is described Community White paper submitted by Mikhalevsky, Sagen, Worecester et al. 2013 to the first Arctic Observing Summit. This paper demonstrates the capability and interest in using acoustics as part of an integrative observing system in the Arctic. The community paper suggests the following recommendations for implementation of a future observing system:
1) A multiyear development and implementation plan of multipurpose acoustic network infrastructure that includes international participation should be completed. This plan should start with one or two cabled nodes like the National Science Foundation Ocean Observatories Initiative (OOI) (http://www.oceanobservatories.org/) Regional Scale Nodes (RSN), and Canadian Neptune (http://www.neptunecanada.com/) perhaps off the North Slope of Alaska into the Beaufort Sea or coordinated with the Svalbard Integrated Observing System from the Fram Strait and eventually into the Eastern Arctic.
2) Further develop reliable low- to mid-frequency sources (50-500 Hz) to support the basin and regional scale tomographic, communications and navigation networks. Multipurpose network of regional sources and receivers on moored and ice-tethered platforms should be deployed and maintained annually. These deployments should be coordinated with icebreaker expeditions and other logistical capacities that are expected to be available every year in the high Arctic.
3) Transfer the technology products and lessons learned on the subsea infrastructure being deployed for the OOI RSN into the Arctic.
4) Integration of acoustic data with oceanographic data from ice-tethered platforms, gliders and ships as well as sea ice data, where spatially integrated measurements from acoustics are used in combination with point measurements and vertical profiles of physical variables. These data can then be used in modelling and data assimilation systems.
The implementation of a sustainable Arctic Observing system, including the acoustic networks will need international funding, coordination and roadmaps. On the European side several roadmap project are planned such as the SIOS- The Svalbard Integrated Earth Observing System (Cynan Ellis-Evans, et al. 2013, http://www.sios-svalbard.org ) and the EPOS - European Plate Observing System (see Kuuvet et al. 2013, http://www.epos-eu.org/). The ACOBAR consortium will integrate and collaborate with those projects as well as with bodies such as EuroGoos.
An environmental assessment study was conducted in the beginning of the project as part of the design study, and it was updated by the end of the project. The results were reported in two Environmental Assessment Reports (EAR1 and EAR2). The field experiments built upon the initial underwater acoustic tomography system installed in Fram Strait during 2008 by the DAMOCLES Integrated Project. A post-experiment acoustic and environmental analysis of sound transmissions has been conducted to confirm the predicted potential for impacts on the marine environment that was documented in EAR1. Based on the actual implementation of the experimental equipment and scientific analysis of the acoustic sources, the experiment did not impact the environment of Fram Strait. The low source level of the acoustic sources, combined with their placement in relationship to Marine Protected Areas (MPAs) and the long intervals between transmissions, precluded the experiment activities from affecting any resources of an MPA. Based on the potential for impacts, it is highly unlikely that the experiment activities affected any marine species listed as endangered or threatened under the U.S. Endangered Species Act nor any marine species that are listed as threatened (vulnerable, endangered, or critically endangered) by the International Union for Conservation of Nature (IUCN). Additionally, is very unlikely that any behavioral takes of marine mammals (under the U.S. Marine Mammal Protection Act) occurred as a result of this experiment. In conclusion, the scientific analysis of the ACOBAR experimental activities indicated that no ethical issues resulted, as the experiment had no potential for causing grave danger to marine mammals potentially occurring in the Fram Strait area.
Results from WP2: Arctic Ocean Observatory
The main activity in WP2 has been to develop, test and implement a system of Acoustic Ice Tethered Platforms (AITPs) which can operate over long time in Arctic ice covers areas. The AITP system is a cluster composed of 4 distinct but identical platforms used for navigating and exchanging data with an acoustic under sea-ice glider. The acoustic sea-ice glider is not a conventional sea glider since it is deprived of any direct links with satellites except for an Argos antenna used in case of emergency and back up for ultimate retrieval. The AITP platforms are fully autonomous over periods of several months. Each platform is composed of a surface unit equipped with a satellite Iridium and a GPS transmitter, an underwater ocean profiler moving along a 1000m long cable suspended vertically under the surface buoy, three underwater acoustic modems for exchanging data between the sea-ice glider, the ocean vertical profiler and the surface unit.
In addition, the ocean vertical profiler is equipped with a CTD sensor for measuring temperature and salinity versus depth from surface down to 1000m depth and a SOFAR-RAFOS low frequency acoustic transmitter/receiver (1560-780hz) for transmitting long range acoustic signals to the sea-ice glider. Each AITP is capable of transmitting and receiving SOFAR-RAFOS acoustic signals several times per day in order to estimate precisely the sound speed between AITPs and the sea-ice glider for navigation and localisation purposes. The underwater vertical ocean profiler is equipped with an hydraulic buoyancy control device composed of two bladers filled up with oil in order to displace the ocean profiler up and down along the 1000m long cable suspended under the surface unit. The ACOBAR AITP cluster is composed of 4 identical platforms operating in a similar way. Each Ocean profiler is profiling once per day from surface down to 800m depth to measure temperature, electrical conductivity and pressure from which one can deduce potential temperature, salinity, density and depth. This profile is transmitted from the ocean profiler to the surface unit via the acoustic modems installed both on the ocean profiler and the surface unit. Then this vertical CTD profile is transmitted via Iridium to satellites in near real time. CTD profiles are then stored on a server at the French Polar Institute (IPEV).
In addition to the CTD profiles, the ocean profiler is equipped with a Seascan system to send SOFAR and to receive RAFOS signals. The system operates once per day and uses 4 dedicated acoustic windows. For each window only one ocean profiler is in sending mode while the 3 other ocean profilers are in receiving mode. Each ocean profiler is configured to use 25 minute window starting at 11h00 each day. The Seascan RAFOS system returns the TOA of the six highest correlation level signals received during one acoustic window. The Seascan SOFAR system uses a 224 samples sweep at a resolution of 0.09230 second per sample and corresponding to a 20 second sweep.
The AITP is transmitting acoustic data from the SOFAR-RAFOS system (TOA and Correlation levels) and some technical data (voltage, pumping actions etc...). TOA concern Time of Arrival of the acoustic SOFAR signals transmitted from one AITP to the 3 other AITP part of the same cluster. One can then compare the travel time between pairs of AITP with GPS positions in order to deduce underwater sound speed. This is the sound speed that will be used for navigating glider under sea-ice. TOA need to be corrected from ocean profilers clock drift. TOA resolution is about 0.1s (equivalent to about 150m). Levels of correlation are giving an indication about the strength of the acoustic signal received by each AITP ocean profilers. In fact the 6 best correlation levels are recorded inside the same window frame. Only correlation levels well above noise level are selected for each TOA.
Readiness of AITP cluster for testing in Arctic Waters is mainly related to the capacity of AITP to garantee a full time operational behaviour by accomplishing (1) the basic transmission schedule both involving the underwater double acoustic system (SOFAR-RAFOS and Acoustic modems) several times per day, (2) the satellite data transmission schedule (Iridium) and GPS positioning several times per day and (3) the daily vertical ocean profiles from surface down to 1000m depth maximum. In order to test it we selected the transpolar drift terminal section from the North Pole to Fram Strait between Spitsbergen and Greenland. Deployments of AITP occurred during Spring time near the North Pole taking advantage of the Russian Barneo station installed near the Pole every year for one month in April. The recovery was organized during the Fall (September-October) from Norwegian vessels operating in the Fram Strait area (Lance and KV Svalbard). The duration of the test was about 5 to 6 months covering Spring and Summer (winter and summer conditions). Two tests were conducted during ACOBAR in 2012 and 2013 respectively over a 6 month period and allowed to detect and fix several technical problems.
Basically the AITP is now proven for operating in real Arctic conditions from winter (freezing) to summer (melting). There are still some improvements to be done for increasing the duration of operations in particular for ocean profilers to profile for 6 months. The actual limitation is now of about 3 months due to several problems for activating the hydraulic pump. Too many pumping actions for each profilers are decreasing the autonomy of the ocean profiler. The ocean profiler is a large underwater neutrally buoyant float with some drag effect that needs to be compensated and generate more pumping actions increasing energy demand significantly. A single rider instead of two for the actual version of the ocean profiler might be able to decrease the drag along the 1000m long cable during each daily vertical profiles. The SOFAR-RAFOS acoustic system needs to be improved for increasing range of propagation for the acoustic signals allowing a better tracking of the under sea-ice glider. Both the strength of the acoustic SOFAR signals and a better precision and resolution of the RAFOS detection (Correlation) are needed. Finally the Iridium transmission and GPS localisation need also to be improved. Thanks to the addition of an Ice Mass Balance (IMB) system on each AITP measuring temperature profiles across sea-ice and equipped with an autonomous Iridium and GPS transmitter, we were able to realize a good tracking for each AITP during the two operations in 2012 and 2013. It was quite fortunate since we met some problems with AITP iridium-GPS transmitters during the two years.
The Ocean profilers are quite reliable and the quality of the data is excellent. The Acoustic modems are also quite reliable and the data transmission between the ocean profilers and the surface units is quite good. The general and overall configuration of the AITP is quite well adapted to operate in the Arctic during freezing and melting conditions. This is facilitating deployments and recoveries as well. The region selected for these experiments extended from the North Pole down to Fram Strait over more than 1000 km range. The second test in 2013 consisted in two AITP platforms deployed in the vicinity of the North Pole in April 2013 on two ice floes located 10 km apart. It took about 6 months (April to September 2012) to drift over more than 1000 km between the North Pole and Fram Strait.
The data collected consists of about 60 vertical profiles of temperature and salinity obtained from both AITPs over a three months period demonstrating the reliability of the whole system. The data shows the presence of a cold subsurface layer at freezing temperatures located at about 100m depth enriched with brines rejected during sea-ice formation in winter. The resulting salinity for this brines enriched layer was about 34 ppt salinity. At greater depths (300m to 400m) the Atlantic layer is quite recognizable with temperatures exceeding +1°C and salinities reaching 34.80psu salinity. Surface salinities were low at the North Pole (30psu) and were increasing further south (34psu salinity).
The underwater ocean profiler, which also includes a dual acoustic system, provides CTD data from surface down to 800m depth on a daily basis. The CTD data are fundamental for observation of water masses and ocean processes under the sea ice. In addition the CTD profiles can be used for estimating the suitable depth for optimizing long range acoustic propagation between AITPs and the sea glider operating under sea-ice. The dual acoustic system operating between the AITPs is composed of long range SOFAR transmitters and RAFOS receivers and short range acoustic modems for data exchange. Therefore the AITP cluster provides essential information to the sea glider in order to navigate under sea ice over long distance (SOFAR-RAFOS) and to exchange data at short range (Modems). The AITP surface unit provides all the necessary information concerning real time positioning of the AITP using GPS. In addition the surface unit communicates with satellites in order to transmit data and exchange information in both directions between land stations and the AITPs.
From a technological point of view, the ACOBAR AITP is now a proven and reliable instrument based on two consecutive tests occurring in real Arctic conditions covering both winter (freezing) and summer ( melting) seasons over a 6 month period and over more than 1000km range. The acoustic modem transmission between the surface unit and the underwater unit was quite reliable. The ocean profilers have completed between 40 and 60 high quality vertical CTD profiles between surface and 800m depth during field tests in 2012 and 2013. The satellite transmission worked perfectly during the second field test (still ongoing) for the two AITPs deployed in the vicinity of the North Pole in April 2013. The surface unit processor (persistor) proved to be quite reliable.
Improvements are needed for expanding the longevity of the system by increasing battery supply and/or reducing the energy consumption. In the actual situation the autonomy of the ocean profiler is limited to 3 months and should be increased by a factor of 2 at least.
The area selected for these two preliminary field tests located between the North Pole and Fram Strait is ideal for launching the first ACOBAR AITP cluster providing long-term navigation capabilities for a sea glider operating under sea-ice. The ice divergence is not creating any serious problem in distorting the AITP cluster in any significant ways.
From a scientific point of view, there is a great interest in exploring the transition zone between the North Pole and Fram Strait which is an important part of the Arctic transpolar sea-ice drift. A sea glider would provide the missing information concerning the surface layer enriched with brines at freezing point and 34psu salinity in some regions that are actually difficult to map. Also a sea glider would provide missing information about the surface layer still fresh in the vicinity of the North Pole and more salty and colder further south heading in the direction of Fram Strait. It seems like there is a frontal region located at about 87°N where salinity is increasing drastically. All CTD profiles provided by AITP ocean profilers, indicated the presence of a well-defined Atlantic layer at about 300m depth. Based on AITP CTD profiles, there was no sign of a double diffusion mechanism occurring above the Atlantic core and within the main thermocline as it was identified in the central Arctic Basin recently. But this might be due to lack of vertical resolution of the AITP CTD profiles. A sea glider could overcome this difficulty.
It was also interesting to observe significant divergent sea-ice motions related to atmospheric forcing with low pressure systems tending to create convergence and high pressure systems creating some divergence in sea-ice dynamics. But on the long term these converging-diverging sea-ice motions did not alter significantly the geometry of the AITP cluster. In addition to the AITP, we deployed IMB (Ice Mass Balance) systems to provide information regarding temperature profiles across the sea-ice and sea-ice thickness evolution as a function of time. But this is not part of this report. However as an integral part of AITP it was interesting and useful to collect basic meteorological data such as air pressures and air temperatures versus time at surface (the meteorological sensors are located within the AITP surface unit). Both measurements proved to be reliable and could be used to complete networking information such as IABP (International Arctic Buoy Project) and to serve for calibrating satellite sensors. For our application it was interesting to document the highly correlated sea-ice diverging motion and surface air pressures transiting from low to high pressures during a single event. It was reassuring and important to observe that phenomenon had no significant impact on the AITP cluster geometry as far as navigating a sea glider under sea-ice with an AITP cluster is concerned.
The next phase would consist in deploying a full AITP cluster in the vicinity of the North Pole in springtime and to let it drift southwards in direction of Fram Strait together with an acoustic sea glider operating inside the AITP cluster. The AITP would provide all the necessary information to the sea glider operating under sea-ice constantly and releasing data to any of the AITP as soon as it would get within the acoustic range for communicating with the acoustic modems. The AITP grid could be of the order of 100km. The experiment would last 6 months and all the equipments could be retrieved in Fram Strait the following September.
Results from WP3: Fram Strait Ocean Observatory
The acoustic experiments
The Fram Strait acoustic network for tomography, navigation and passive acoustic were deployed in 2010. The system was recovered in 2012. Data, including acoustic recordings and engineering data, from the recovered instruments have been scanned and processed. The recovered instruments in A, B and D provide full year acoustic receptions from 16 hydrophones and engineering data required for clock correction and mooring motion correction. In total data from five hydrophone arrays each consisting of 4 hydrophones provided acoustic data. Except for the first deployment in D (2010-2011) all the arrays had a 9.6 meter spacing between their hydrophones. This spacing allowed for beam forming.
After the engineering data was processed, the sound recordings are corrected for mooring motion and clock drift. The recordings are then analyzed with the aim of identifying ray arrivals of the frequency sweep signals emitted by the sound sources. Matched filtering is used to compress the 1-minute sweep signal, thus optimizing the signal to noise ratio. Beam forming provides information about the arrival angle of the wave front. The estimator-correlator accounts for the incoherence of recorded sweep by using sensible estimates of the coherency bandwidths of the received signal in time and frequency. The data processing in ACOBAR is a clear improvement from the DAMOCLES. The improvement is partly due to the experiment design, which had a spacing of 96 m.
The processing described above was carried out for the ACOBAR data and the time series of travel times and arrival angle along 3 sections has been produced. The first set of travel times was forwarded to WP 4 for inversion to temperature in March 2013. This was later than planned, due to several issues that had to be solved before the data were useful for inversion.
To summarize:
1) The positioning of the transponders and anchor points before the mooring corrections can be made is more complex (more like a puzzle) when the positioning is done from a ship where the interrogating transducer is not fixed to the ship, but hanging over the ship side. Also, the GPS positions given relative to the ship antenna need to be corrected to provide the position of the transducer. Finally, “cleaning of the positioning data” is required to get rid of the data, which are not correct. This is tedious and time consuming work.
2) The most serious problem was that the source functions describing the signal transmitted from AWI and NERSC 1 sources did not follow the specification provided by the “source producer”. When the source signal is not accurately known this leads to problems in the pulse compression and the result is smeared arrival peaks, which results in larger errors in the exact detection of the arrivals. The smeared out arrivals was observed in the DAMOCLES experiment, but it was not known why it was like this. The deviation from the specification was realized after calibration of similar sources owned by Scripps Institution of Oceanography. There is a timing error of up to 50 ms. To get an exact description of the source function the sources should be properly calibrated, but this is expensive. This will be followed up in later projects if funding allows. Otherwise a synthetic description of the signal will be formulated using measurements in the Laboratory.
3) The oceanographic profiles in the open ocean part of the Fram Strait makes the arrivals come almost at the same time like a cluster of un-resolvable arrivals. A significant effort was carried out to optimize the signal processing.
The C mooring is not yet recovered and dredging for the acoustic source in mooring C will be performed as part of the Norwegian funded UNDER-ICE in 2014. All transponders around C mooring were recovered.
Use of gliders and oceanographic moorings
A total of eight glider missions were carried out during the ACOBAR period, both in summer and autumn seasons. During each mission the measured data as well as engineering data of the mission were transmitted in the near-real time to the base station operated by Optimare in Bremerhaven. Gliders were deployed typically for 2-3 months with turn around the vehicles in September, usually during KV Svalbard autumn cruises. All summer deployments were successful with gliders recovered after the end of a mission while during autumn deployments three gliders were lost after between 1 and 2 months due to technical problems (broken antenna or battery failure). However, also prematurely finished autumn missions delivered significant number of hydrographic profiles and RAFOS receptions. Gliders were operating in the eastern and central Fram Strait, mostly in open waters and at the ice edge.
Developing operational capability for the near-real time (NRT) data transfer from the moored array was focused on two main aims: to test the long-range acoustic data transfer between moorings and to develop the communication mooring with the surface unit capable of satellite data transmission. To achieve the first goal, three low-frequency ling-range acoustic modems, the HAM.nodes were tested in Fram Strait in different configurations. Since acoustic data transmission over a typical range between moorings of O(30 km) proved to be unreliable, the distance between long-range modems was reduced by adding a relay-link mooring with additional modem in a half-way between instrumented moorings.
ACOBAR deployments of acoustic modems revealed severe difficulties in achieving a long range communication between the Fram Strait moorings over a whole deployment season. Operation time of the units in many cases was limited due to system failures and thus the tests were done over shorter periods. Moorings equipped with acoustic modems were located in highly energetic area with strong currents, in consequence moorings were rarely in vertical positions due to strong diving. Environmental conditions and a depth of sound channel varied significantly between seasons and signal to noise ratio was relatively low in most cases. The modem technology was improved in the meantime but several options were still lacking during the ACOBAR period. The large number of packets per telegram led to the loss of several datasets during sending attempts during the first long-term deployment. The ACOBAR modem deployments can be concluded the horizontal long-range communication might be possible over a range of about 10 km in Fram Strait if modems could be held in one depth with only little variation. Due to strong variability of environmental conditions, handling of transmission schedule should be improved by checking sound channel properties and depth information first before transmissions. The model technology is promising but still not mature enough for data transmission over the ranges longer than 10 km.
The NRT data transfer from moorings to the land station was addressed by a design of the field experiment with two different systems. The first system consisted of the underwater winch which was equipped with the two different profiling CTD tops: one built on a base of the NEMO float and one being the generic NGK profiler. The stand-alone mooring with a tethered standard NEMO float was also tested but it was found that it could not negotiate the strong currents therefore was not able to reach the surface. The profiling winches used for test deployments of several weeks worked reliably according to the programmed schedule. However, the first tests of the NEMO-based CTD profiler in 2010-2012 were not successful. The profiler either got lost or was damaged (flooded or broken) during three subsequent deployments and no communication was established with the land station. Only during the last deployment in 2012, the heavy instrumented NGK profiler provided 4 profiles per day (213 in total) and reached the surface during nearly each ascension. Data were also partially sent to the land station. Therefore, based on ACOBAR output, this technology will be used for future development of NRT data transfer from the Fram Strait moored observatory.
The NRT data transfer was also achieved during the glider missions in Fram Strait in 2010-2012. During all missions, the full engineering and scientific data sets from the upper 1000m layer were transmitted by the glider during surfacing after completion of each single dive. The temperature, salinity, dissolved oxygen and fluorescence profiles and sections were processed and visualized in the near-real time during the on-going glider mission. There is on-going development focused on application of a short-range acoustic modem dedicated to data transmission between glider and underwater moored instrumentation. Therefore, future solutions for using gliders as data messengers from a moored array are considered as more robust then currently tested moored profiling solutions.
To implement the acoustic positioning and navigation system for gliders and floats for future NRT data transfer from under the ice, three RAFOS sources with frequency 260 Hz were deployed in July 2010, four RAFOS sources in July 2011 and seven in July 2012. Additionally three tomographic sources, deployed in 2010 (and partially redeployed in 2011) in a triangle covering the northern Fram Strait were programmed to provide the RAFOS transmission between tomographic sweeps. During all missions, gliders collected RAFOS receptions and calculated navigational solution based on RAFOS signal using the built-in RAFOS hardware and the dedicated firmware developed by the APL-UW group led by Craig Lee. In summer 2010 the glider received 447 RAFOS transmission with maximum ranges for RAFOS source located in the ice-covered area up to 250 km. During the summer mission in 2011 the glider collected altogether 298 valid RAFOS receptions, used to calculate 68 positions, of which 70% with error less than 10km and 25% with error less than 5 km. The accuracy of positioning improved for gliders equipped with a new type of RAFOS hardware when during the autumn mission in 2011 72 % of 99 reliable positions had an error less than 10 km and 40% less than 5 km. 147 valid positions were based on 483 acoustic receptions collected by the glider in summer 2012 (40% with error < 10km). The accuracy of acoustic navigation is acceptable for under ice mission but can be still improved by using more accurate clocks in sound sources and better synchronisation between sources and glider RAFOS hardware.
Results from WP4: Assimilation of tomography data
The aim of the work in WP4 is to invert acoustic travel time data to temperature fields and develop of assimilation schemes for use of acoustic travel time data in ocean models. The acoustic measurements are different from other oceanographic measurements by its integrative characteristics. The challenges of the assimilation schemes are to understand how the wide arrival pulse reflect, how to quantify the distance between the observed broad pulse with the ray predicted arrivals, and how can we cope with this inversion and assimilation.
Acoustic travel time measurements are inverted to range averaged sound speed profile which again is averaged over depth to a range-depth averaged sound speed to achieve a very high accuracy. A simple scale factor of 4.5 m/s/OC can be used for conversion of sound speed variations to temperature variances in the Fram Strait. The accuracy of the conversion is within a few percentages. In the case of assimilation the temperature and salinity will be updated within its dynamical constraints.
A numerical investigation show that the shape of the single, broad arrival pulse of around 100 ms width arises from a combination of the minimal dispersal of the acoustic arrival pattern cased by the fairly weak sound speed channel in the Fram Strait and internal waves causing scattering of the acoustic field. The internal wave scattering results in proliferation of eigenrays that essentially fill the water column between the source and receiver – this indicate that the individual ray paths no longer have a meaning. This points towards the use of Time Sensitive Kernel as described by Skarsoullis et al. 2009 which has not yet been used for inversions. This methodology can also create machinery for future assimilation. This approach will be addressed in upcoming project proposals.
The inversion scheme developed is based on vertical EOFs established based on the high-resolution ice ocean model developed within the DAMOCLES. The horizontal variability is handled by sinus and cosines. Most important, the principle of identification of specific arrivals with specific rays were abandoned the. The strategy was to quantify the oceanic variability apparent in acoustic arrival patterns consisting of wide acoustic pulses of some 100 ms in width. Since all computed rays arrive at the same time, it does not matter which ray is assigned to which particular peak, so first order. Only the peaks from the single-wide pulse were employed for this inverse. Large 100 ms uncertainties were assumed for the data, however, to account for the lack of ray identifications. The inverse problem is saved by the essential resolution of the integrating measurements, however. Furthermore, the large amount of data minimizes the impact of the large data uncertainty (the data being picked-peak travel times).
Ultimately, the formal uncertainties in temperature estimated from the inverse were about 50m°C. The high-frequency variability of the temperature time series was larger, with a 78m°C RMS. The corrections to the Fram Strait model in sound speed were then used to estimate absolute temperature, using a conversion factor of 4.5 m/s/°C. The estimated temperature variations were in remarkable agreement with earlier estimates by Skarsoulis et al. 2010, who used a completely different inverse approach. In future experiment it may be that employing a deep receiver at 1700 m may result in a stable resolved ray which may double the available information for the inversion.
The inversion method is used in a numerical study to compare the use of point measurements and tomographich measurements either separately or combined. It is shown that for obtaining an estimate of an average of a scalar quantity along a section across Fram Strait, optimal measurements are a combination of point and line-integral data types. This combination reduced the uncertainty of the estimate by about a factor of three. An equivalent reduction of uncertainty using only point measurements would require a roughly order of magnitude increase in the number of point measurements. This result demonstrates the fundamental motivation behind employing acoustic tomography in conjunction with other data types.
Our hypothesis of ACOBAR is that the tomography data will improve or constrain the model estimates for temperature within Fram Strait, hence lead to a more accurate picture of the exchange of heat between the Atlantic and Arctic Oceans. As can be seen in the sections below, considerable progress has been made toward this goal.
The work to implement data assimilation with line-integral constraints (tomography) is technically challenging. Starting point of the work in ACOBAR is based on the results from the single track experiment conducted in DAMOCLES from 2008 to 2009.. The oceanographic conditions in the Fram Strait cause the acoustic arrivals to come in a cluster. This makes it impossible to use methods based on separation of individual arrivals. This influences how to establish the assimilation and inversions schemes. In DAMOCLES this initiated an analysis of how to establish the measurement matrix and how to compare the modeled arrivals with the observations. The approach of identifying and comparing individual rays were abandoned in both the assimilation and the inversion approach (Skarsoullis, et al 2010). In the assimilation another problem was encountered. The modeled observation matrix (predicted arrivals) did not match the acoustic observations either in the size of variability or the seasonal variability (Haugen et al. 2010, Sagen and Sandven, 2006, Bertino et al. 2009, Sagen et al. 2010 a,b, Haugen et al. 2010). The ice-ocean model system developed within DAMOCLES for the Fram Strait, and status in the use of Ensemble Kalman filtering with acoustic measurements is provided by Sagen et al. (2010 a,b).
The major findings in the experiments was that
1) Acoustic arrivals in the West Spitzbergen Current come in as a broad pulse of un-resolvable arrivals making it impossible to detect arrivals and identify a ray arrival to it.
2) A “brute force” inversion based ray arrivals was developed and provided depth-range mean temperatures
3) Modeled observation matrix were discussed and compared to the observations and used as a tool to validate the high resolution ice-ocean model and the Topaz 3 model.
4) Distance between modeled and observed acoustic travel times were difficult to quantify due to point 1) and also because the ice-ocean models are not representing the ocean good enough and the modeled observation matrix is strongly dependent on the interpolation routines in both the vertical and the horizontal.
The activities in ACOBAR has first of all aimed towards providing new, and essential ground work for data assimilation. In this first phase of WP 4 the travel time data originated from the 2008-2009 experiment has been used (Haugen et al. 2010). This was because acoustic data was only available after the cruise in 2012. In the work described in this report the nature of the unique acoustic properties within Fram Strait has been further elucidated, preliminary inversions of the acoustic data to obtain estimates of temperature have been completed, and the expected impact of the acoustic data compared to such data as the set of Moored Array thermistors has been quantified. We have a much better understanding of the expected impact of these data on the numerical model, hence how to test the integrity of the eventual data assimilation results.
The second issue has been that it has been difficult to precisely quantify the differences between ocean model predictions and observations such that these differences can be used to correct, or constrain, the ocean model. The discussion and resolution of this latter issue comprises a large part of this report.
Results from WP5: Dissemination and exploitation
The ACOBAR project website (http://acobar.nesc.no) has been established and populated with information about activities in the project. The other participants have contributed with various dissemination material for the website. AWI has established a dedicated website for glider activities in Fram Strait (http://www.awi.de/?id=6021). It contains background information on gliders, technical description of glider types used in ACOBAR (with a special focus on Seagliders used in Fram Strait), a summary of state-of-art regarding the acoustically navigated gliders and a description of the planned multipurpose acoustic system for glider and float navigation in Fram Strait. The picture gallery and a collection of relevant publications and links related to different glider activities are also provided.
The first version of the dissemination and use plan was provided in period 1 (D5.2). An updated version of this plan has been prepared at the end of the project.
Dissemination activities include publication of peer review papers in science journals, conference proceeding, poster presentations, oral presentations and meetings with users. A detailed list is presented in Table A2.
Links to other projects developing underwater observing systems. ACOBAR has been presented to the ESONET Network of Excellence. ACOBAR will contribute to ESONET by an Arctic demonstration project in 2010-2011. ACOBAR has also established links to Norwegian Defense Research Establishment, Kongsberg Maritime and other companies working with AUVs and underwater acoustic systems
Results from WP6: Transfer of technology and knowledge
Activities at NERSC
The transfer of technology and knowledge in ocean acoustics has been obtained through close collaboration with the partners Scripps Institution of Oceanography (SIO) and Woods Hole Oceanographic Institution (WHOI) in USA. The activities have included exchange visits between Norway and USA, training in use of acoustical systems, planning and implementation of field experiments, data processing, presentation at workshops and conferences, and publication of papers. Students and young scientists have been involved. The activities started in the DAMOCLES project (2005-2009) and has continued as part of ACOBAR from 2008.
In April 2009 Hanne Sagen and Stein Sandven from NERSC visited WHOI and Webb Research Coorporation as part of the preparation of the acoustic tomography system. The visit was combined with an ACOBAR progress meeting (14 April) and a Workshop (15 April). In August 2009 Dr. Peter Worcester from SIO visited NERSC in Bergen and Longyearbyen and participated in the cruise with RV Håkon Mosby. The tomography moorings were recovered and one year of tomography data were obtained. Dr. Peter Worcester gave a guest lecture on ocean acoustics when he was in Bergen.
Hanne Sagen and Svein Arild Haugen from NERSC visited Scripps Institute of Oceanography (SIO), USA for 14 days (22 October - 6 November 2009). The aim of the visit was to process and start analyzing the acoustic data from the first Fram Strait experiment 2008-2009. Data analysis was completed after returning to Norway. Svein Arild Haugen used this work in his Master thesis completed in April 2010.
• The cooperation with WHOI and Scripps has resulted in several publications and reports
• Advanced acoustic receiver and source technology has been purchased from Scripps Institution of Oceanography, and Teledyne Webb Research Cooperation, respectively. NERSC personnel have been trained in deployment, maintenance, testing and deployment procedures by engineers and scientists at the SIO. SIO have participated in the field operations and they have worked closely with scientists form NERSC, University in Bergen (UIB), and AWI.
• The most efficient transfer of knowledge takes place during planning and execution of field experiments. Peter Worcester and Andrey Morozow, WHOI/Webb Research participated in the RV Håkon Mosby first deployment cruise in 2010. Lora Van Uffelen (Post Doc) from SIO and Peter Koski (Engineer) from WHOI (communication) took part in the KV Svalbard Cruise (2010). Peter Worcester and Lloyd Green participated in the KV Svalbard cruise, 2011 (recovery/redeployment). Aquatec Ldt (2009, 2011) and NAXYS (Norwegian SME) (2008, 2009, 2010) have participated in several cruises. ENSTA and NKE participated this year in Håkon Mosby cruise.
• Brian Dushaw, Applied Physics Laboratories, Seattle, is visiting scientist, funded by Fullbright fellowship and ACOBAR, from 1 September till end of February. Dushaw has long experience in inversion of acoustic, and he is working together with Hanne Sagen on assimilation/model oriented inversion in WP 4. Dushaw took part in the field experiment in 2011 and 2012.
• Scientific visits from SIO and WHOI to NERSC: Dr. Peter Worcester and Mathew Dzieciuch
Activities at AWI
Regarding use of gliders in Arctic waters. AWI has cooperated with Applied Physics Laboratory, University of Washington, USA (APL-UW). AWI uses the Seaglider produced by APL-UW, and the first successful testing of this glider in the Fram Strait was done in 2008. An adaptation of the Seaglider for under-ice operations has been done under a collaboration with the Integrative Observational Platforms Group (APL-UW) led by Craig Lee. The first winter long Seaglider deployment in Davis Strait from September 2008 to February 2009, performed under the US National Science Foundation project 'An Innovative Observational Network for Critical Arctic Gateways'http://iop.apl.washington.edu/projects/ds/html/program.html proved feasibility of extended under-ice mission where Seagliders navigate using acoustic signals received from moored RAFOS sound sources. In ACOBAR, both conventional RAFOS sound sources and tomographic sources will provide RAFOS navigation signals and the RAFOS technology provided by the APL-UW will be used for under-ice navigation of Seagliders in Fram Strait. OPTIMARE acts as the link between Seaglider manufacturers and the end users (scientists) to promote the exchange of information and requirements and to assist the field operations.
An adaptation of the under-ice Seaglider and acoustic navigation technology used in Davis Strait for Fram Strait has been done under a collaboration with the Integrative Observational Platforms Group (APL-UW) led by Craig Lee. The first winter long Seaglider deployment in Davis Strait from September 2008 to February 2009, performed under the US National Science Foundation project 'An Innovative Observational Network for Critical Arctic Gateways'http://iop.apl.washington.edu/projects/ds/html/program.html proved feasibility of extended under-ice mission where Seagliders navigate using acoustic signals received from moored RAFOS sound sources. In ACOBAR, both conventional RAFOS sound sources and tomographic sources will provide RAFOS navigation signals and the RAFOS technology provided by the APL-UW was used for acoustic positioning and navigation of Seagliders in Fram Strait. In collaboration with the APL-UW group the system used in Davis Strait was adapted for Fram Strait, taking into account the environmental conditions and different frequency of available RAFOS sources.
Activities at UPMC, ENK, ENSIETA and AQUATEC
There has been extensive transfer of technologies between UPMC, ENK, Aquatec and ENSIETA on the development and testing of the AITPOPS cluster with glider. ENSIETA has been recently exchanging data and knowledge with AWI, this cooperation helps ENSIETA to design its long range positioning algorithms using AWI's RAFOS data acquired in the Fram Strait
Activities at ACSA
The expertise of ACSA in development of underwater navigation system has been used in analysis of the glider navigation data obtained during the experiments in the Fram Strait
Potential Impact:
ACOBAR has developed and demonstrated sensors and platforms that can improve the in situ observing network considerably compared to the present situation. The ocean observing system provided by ACOBAR is expected to have impact on a research and monitoring of the polar regions by improving the in situ observations of physical and biogeochemical parameters of the marine environment. Several thematic areas will benefit from ACOBAR such as ocean and climate monitoring, marine transportation and safety of operations, ocean acoustics including navigation and communication, oil and gas exploration, marine resource management, and protection of marine ecosystems in the Arctic. Development and implementation of ocean observing system for this wide range of user groups requires active promotion and dissemination of the results from ACOBAR after the project is completed. The need to enhance the in‐situ marine monitoring capacity by on global scale has been recognized by GOOS, GCOS and others, while on European scale EuroGOOS is the main coordinating body for operational ocean monitoring and forecasting. All the 34 EuroGOOS members from 16 countries play key roles in improvements of the in‐situ marine monitoring in the European seas including the Arctic. One condition for improvements of the observing systems is, among others, new investments in RTD activities that would deliver new, more efficient, robust and cost effective sensors to equip different platforms such as buoys, gliders and floats. The ACOBAR project has therefore demonstrated a significant improvement of the ocean observing systems in the Arctic seas. The next step after demonstrating the capacity of the ACOBAR system is to implement a larger number of platforms that can operate over large parts of the Arctic Ocean. This requires cooperation with USA, Canada, Russia and other nations to install and maintain a network of platforms in the Arctic. It also requires that research and monitoring institutions in these countries increase their funding of long-term Arctic programmes that can sustain the monitoring system.
Multinational efforts are required to build an ocean observing system in the Arctic based on a network of in situ sensors and platforms. It is necessary that European institutions cooperate with USA, Canada, Russia and other countries in order to implement and sustain a pan-Arctic ocean observing system.
Increased availability of standardised in-situ data
ACOBAR has provided oceanographical data from the AITPs, the gliders and the other platforms which are interoperable with other data repositories for oceanographic data. The data collected are delivered in standard formats established by the oceanographic community. These formats are used in operational observing systems such as the ARGO programme (http://www.argo.ucsd.edu/) and are adopted in many European projects (SeaDataNet, EMODNet, MyOcean, GISC). SeaDataNet is coordinating European ocean and marine data management, and defines standards and software for implementation of this work (http://www.seadatanet.org/Standards-Software).
Data from Ice-tethered profilers (ITPs) have been established in the Arctic by WHOI almost 10 years ago, and data from these profilers are provided in standard formats (http://www.whoi.edu/page.do?pid=20781). The AITPs in ACOBAR provide data in the same formats. At present data from ITPs and gliders are provided as profiles to the GMES Marine Core Service operated by MyOcean & MyOcean2 FP7 projects using some agreed processing methods developed within FP5-MFSTEP, FP6-MERSEA and FP7_MyOcean projects. In the GROOM project, a data management plan is under development where a single access point for all European gliders will be established. The glider data from ACOBAR will be included in this plan.
Glider data from ACOBAR will also be linked to the OceanSITES program, which is the global network of open-ocean sustained time series sites, called ocean reference stations, being implemented by an international partnership of researchers (http://www.oceansites.org.). OceanSITES provides fixed-point time series of various physical, biogeochemical, and atmospheric variables at different locations around the globe, from the atmosphere and sea surface to the seafloor. The program’s objective is to build and maintain a multidisciplinary global network for a broad range of research and operational applications including climate, carbon, and ecosystem variability and forecasting and ocean state validation.
Establishment of web-interface to marine data repositories is now standard services provided by many data producing institutions. In Europe several ongoing project are developing web-based systems for marine data. NERSC is coordinating the NETMAR project (Open Service Network for Marine Environmental Data), which develops tools for searching, downloading and integrating satellite, in situ and model data from ocean and coastal areas. It is a user-configurable system offering flexible service discovery, access and chaining facilities using OGC, OPeNDAP and W3C standards. Data from ACOBAR will be provided in web-system that is fully compatible with other European projects (i.e. NETMAR, SeaDataNet, EMODNET, etc.).
Reduced cost of data collection system in support of fisheries management;
There is a great need to develop and improve the temporal and spatial scale of in situ oceanographic observations, and this requires more cost-efficient observing systems to be operated from many different platforms (ships, floats, gliders, ice buoys, etc.). One of the goals of ACOBAR was to develop and demonstrate more cost-effective systems for use in the Arctic. CTD instruments (Conductivity, Temperature and Depth/pressure) are fundamental for observation of physical, biological and biogeochemical ocean parameters. They are measured in their own right to understand ocean processes such as water mass movement, but are also required to give physical context to a huge variety of other measurements. Acoustic methods require sound speed profiles obtained by CTD data. As these parameters are so crucial, they are likely to feature extensively in the MSFD monitoring requirements. From an operational perspective, these basic observations are required as standard across a broad range of applications from climate to fisheries, shipping, and energy exploitation.
CTDs and other sensor packages must be operated from platforms that can improve the temporal and spatial sampling of data. In ACOBAR, this was demonstrated by using ice platforms in combination with glider, which can be operated autonomously over long time. This will be the most cost efficient method for data collection in the Arctic Ocean. In the MIZ and open water areas, which are most important for marine resources and fishery management, gliders can in principle be operated year round with data transmission when the gliders are at the surface in open water. Continuous monitoring of physical, biological and biogeochemical will provide more data to support fisheries management. Ship surveys will still be needed at certain times of the year for water sample collection and other measurements that cannot be done with automated sensors. But, gliders and ice buoys with profiling instruments can take over much of the regular data collection from sensor that can operate automatically over long time. A European infrastructure project (GROOM) has been established to develop use of underwater gliders for collecting oceanographic data (http://www.groom-fp7.eu/doku.php). This new infrastructure shall be beneficial for a large number of marine activities and societal applications, which can be related to climate change, marine ecosystems, resources, or security and which rely on academic oceanographic research and/or operational oceanography systems. Several partners in ACOBAR are also members of GROOM (UPMC, AWI, NERSC).
Advance competitiveness for European Industry's & particularly SME's within the Marine sensing sector;
Companies representing offshore industry, underwater technology and manufacturer of observing systems have been involved in ACOBAR both as partners, subcontractors and users. These companies will be the key drivers for developing observing systems to be used in the Arctic in the future. As an example, Scripps Institution of Oceanography will provide expertise and technology in underwater acoustics. Furthermore, Teledyne Webb Reserch, which is a world leading manufacturer of gliders and sensors for use on gliders, will contribute to build up European expertise in this technology. Users of the observing systems will provide useful feedback to the companies based on the testing of the systems in the field trials. The experience gained from ACOBAR will strengthen the competitiveness of the European companies in the marine observing sector. Experience from the Arctic field trials will be a competitive advantage for companies planning to be in the forefront of Arctic underwater technology.
Enable better cooperation between key sectors (Manufacturing Industry, ICT, Maritime Industry, Marine Science, Fisheries etc.);
ACOBAR results will continue to be disseminated to representatives from the manufacturing industry (Teledyne Webb Research), maritime industry / shipping working with focus on the Arctic (Norwegian Coastal Directorate, Centre for High North Logistics), marine science (representative from EuroGOOS), fisheries (Norwegian Fisheries Directorate) and ICT (SeaDataNet, NETMAR). The technologies and results from ACOBAR are of interest for all theses sectors because of the increased capacity to collect in situ data. It is envisaged that better cooperation between the sectors will emerge since they will have benefit from the same technologies and their data producing capabilites.
Data dissemination from ACOBAR will be interoperable with other European initiatives to coordinate and sustain collection, standardization, archiving and dissemination of marine data, in particular SeaDataNet, EMODNET and GISC. Data from ACOBAR will for example be used by the Arctic Marine Forecasting Centre, which is the Arctic Node of the Marine Core services provided by MyOcean (http://myocean.met.no/). ACOBAR will contribute to build up observation technology and improve data collection which is relevant for all these sectors.
Support implementation of European Maritime Policies (MSFD, CFP, IMP, etc.);
The European Maritime policies aims to provide the mechanisms for clean and healthy marine environment, and ensure that exploitation of marine resources is done in a sustainable way. This is particularly important in the Arctic where exploitation of energy and other resources are expected to grow in the future. The policies are also focused on adaptation to climate change, which has significant impact on environment and ecosystems. This is also particularly important in the Arctic Ocean where global warming is most pronounced and the impact on environment is significant.
The Marine Strategy Framework Directive (MSFD) includes an initiative for the development and actual implementation of an overarching European Marine Observation and Data Network (EMODNet). EMODNet is considered as a network of existing and developing European observation systems, linked by a data management structure covering all European coastal waters, shelf seas and surrounding ocean basins. EMODNet is coordinated at EU level by a task team lead by EU DG MARE and tuned with other European directives (MSFD, INSPIRE) and large-scale framework programmes on European and global scales (GMES and GEOSS), that urge access to, and exchange of, environmental data and information. New regulations from MSFD require observations to determine good environmental status. In addition, the Common Fisheries Policy and regulations regarding noise also need data. Therefore ACOBAR results regarding use passive acoustic sensors to monitor the ambient noise and how it varies with space and time. There are operational demands as well, with fisheries, renewable energy, oil and gas, weather and ice forecasting, shipping and defence all requiring in situ observations to support daily operations as well as long-term planning. Large-scale observations will increase understanding of ocean processes, biology, biogeochemistry, climate change and marine health, as well as improving validation of models. The success of the ARGO programme has shown that large scale in situ observations are possible and hugely beneficial..
Safety of marine operations is one of the pillars of European Maritime Policies, and thus a major reason for building up the GMES marine services (http://www.myocean.eu/web/72-marine-safety-description.php). These services provide large-scale ocean and sea ice monitoring from satellite data and short-term ice-ocean forecasting for the whole Arctic with about 20 km grid cells. The forecasting products are delivered by the TOPAZ system at NERSC (http://topaz.nersc.no) and operated by Norwegian Meteorological Institute through the Arctic Marine Forecasting Center (http://myocean.met.no/). The ocean modelling and forecasting systems require improved in situ observing network in order to validate the models and assimilate data into the models. Data from ACOBAR will therefore be a valuable contribution to the Arctic Marine Forecasting Center, in particular the improved access to near real time data
ACOBAR has demonstrated technologies for in situ data collection which is required for development of ocean and sea ice monitoring and forecasting services. These services are needed to ensure safe and economic marine operations in polar regions where sea ice and icebergs represent a significant risk factor for all types of operations.
Assumptions and external factors affecting the impact
The overarching assumption is that there will be a growing demand for in situ ocean data from the Arctic Ocean. There are a number of reasons for this:
(1) There is a need for more physical oceanographic data in the Arctic because the ocean and sea ice plays an important role in the climate system, and this role is not well understood. The ocean and sea ice models, including climate model, need more data to improve the models and validate the model results.
(2) In addition to physical data there is also need for more biochemical data to understand how the ecosystems work, how the marine resources develope under a changing climate, and how human activities affect the ocean environment.
(3) Growing industrial and shipping activities will have demand for monitoring and forecasting systems, which will require observational data
(4) Satellite earth observation, especially the GMES programme, requires more in situ data in the polar oceans for validation of satellite retrieved data and provide subsurface and deep sea data that cannot be obtained from satellites.
(5) Satellite communication will have impact on how much data can be transmitted from various observing systems to the users. It is assumed that such services will improve and grow fast, offering more and more capacity to transfer data
(6) The growth in industrial activities in the Arctic will require development of underwater technology, and companies working with such technologies are the backbone for building new in situ obsvering systems. The strength of this industry is an important factor to push development an dimplementtaion of new obervsing systems
(7) Arctic marine transportation is expected to grow in the future as a result of reduced sea ice. This transportation is driven by the global market and the comparative advantage of the Arctic transporation routes compared to other routes.
(8) The demand for oil and gas exploitation from Arctic regions will be a main driver for future activitites and will define among others the extent to which Arctic industry, shipping and underwater technology will be developed.
(9) Research and monitoring activities undertaken by various counties working in the Arctic will have impact on the demand for the observing systems developed in ACOBAR.
(10) Geopolitical situation and legal framework will also have impact on the interest of different countries and industries to be present and make investments in the Arctic.
List of Websites:
Public website: http://acobar.nersc.no
Contact details:
Prof. Stein Sandven
Director
Nansen Environmental and Remote Sensing Center (NERSC)
Tel: +47 55 20 58 00
Fax: +47 55 20 58 01
E-mail: stein.sandven@nersc.no
