Core Theme 5: Technological Advancements for Improved near-realtime data transmission and Coupled Ocean-Atmosphere Data Assimilation
The last decade has seen substantial developments in both refined and focussed observations in key locations for THC variability and coupled ocean-atmosphere models. However, two challenging areas have been identified, that require significant technological advancements: Near-real time data transmission from moored observatories and coupled ocean-atmosphere data assimilation. Science outputs of this core theme will feed into an overall assessment of priorities for the development of observing systems and models, required for a quasi-operational decadal THC prediction system (Deliverable D6.1.5). This will be one of the major scientific and technological outputs og THOR.
Lead: M. Visbeck (IFM-GEOMAR); co-lead: Detlef Stammer (UHAM)
Participants: UiB, IFM-GEOMAR, UHAM, NERC-NERC
WP 5.1 Development of near real time data transmission for moored observatories
Today ocean and atmospheric scientists have access to a large range of real-time environmental data. Most of the space-based information is available within hours, and even an acceptable level of quality control is performed within a few weeks. Likewise an increasing number of in-situ data from the ocean are available for data assimilation on similar time scale. The situation is much worse for many moored applications. Notable exceptions are the moored tropical arrays of the TAO/TRITON and PIRATA programs that have significant surface expressions that enable full real-time data transition and subsurface moorings with much smaller surface buoys that have allowed to transmit data in real-time. These solutions are, however, sensitive to fisheries and other disturbances at the surface
In WP5.1, two different systems will be implemented and tested. Neither of them has full real-time capacity, since data transmission, typically, will occur at intervals of a few months. This will, however, imply great advances, both in data timing and security and in economy, since it allows longer deployment times for moorings. The two systems are considered separate tasks of this work package:
Task 5.1.1 Designing and testing a data transmission system using a data shuttle pop-up method, the Kiel-system
This system will be developed based on a pop-up buoy data shuttle technology. A SME (Optimare) has just developed a first prototype of a data shuttle system that can be merged with existing observatory technology. Specifically, a subset of the data from the moored systems will be communicated along the mooring wire to the pop-up buoy control unit. After 4 month of data collection a pop-up bouy will be released and rises to the sea surface. From there the data are transmitted via the Iridium Satellite phone network to shore. Thus the time varying information from several critical regions will be available for data initialisation and/or early verification at a much faster delay time. Today the typical delay time is 1-3 years depending on the mooring-servicing schedule. The specific tasks are:
• Specification of the pop-up system and integration into the mooring system.
• Laboratory and shallow water sea trial in the Kiel Fjord.
• Deployment of the pop-up system in the central and boundary current system of the Labrador Sea.
Task 5.1.2 Designing and testing a data transmission system using acoustic transmission to bypassing ships of opportunity, the Bergen-system
This system does not have any physical contact with the surface during deployment, but will rely on acoustic transmission of acquired data to vessels (ships of opportunity) that pass close to the mooring. The system will be developed by Aanderaa Data Instruments in cooperation with UiB. The moorings will be constructed so that they only need to be recovered, serviced, and redeployed at ~5 year intervals and will comprise bottom mounted ADCPs and high precision CT sensors. The system will be tested by a year-long deployment in the Faroe Bank Channel, where the FFL has regular cruises that will pass by the mooring at least 4 times a year and can upload data.
Lead: Martin Visbeck (IFM-GEOMAR)
Participants: IFM-GEOMAR, UiB, Aanderaa
WP 5.2 Development of coupled ocean-atmosphere assimilation capabilities
Over the last decade significant progress was achieved in ocean data assimilation and data synthesis approaches in the ocean (sometimes also referred to as “reanalysis”). As a result, routine ocean synthesis are now feasible on a routine basis and are being performed over up to several decades, e.g., covering the period of the NCEP/NCAR reanalysis since the early 50th. Those results are now being used to analyse ocean variability and its interaction with the atmosphere. Results are likewise being used to investigate THC and sea level changes of the Atlantic Ocean during the last 50 years (see WP 2). The ultimate use of ocean syntheses is, however the initialization of coupled models. First such attempts were performed and highlighted already the problems involved with improving ocean forecasts by using ocean data syntheses. Some of those problems will be addressed in WP 4.2. However, a fundamental obstacle in reaching progress appears in the fact that ocean syntheses are being performed with a different model than that used during coupled experiments. To improve forecasts data assimilation have to be build into the coupled models that are being used later forcing forecast studies.
This WP 5.2 is, therefore, concerned with improving coupled model forecasts skill using ocean data and ocean synthesis. This includes (1) the improved initialisation of coupled models using ocean syntheses, (2) the evaluation of the improved skill of those coupled models and (3) the building of coupled assimilation capabilities that ultimately will allow, to constrain coupled model directly through climate observations The work will entail several stages that each is of importance and relevance for WP 4. In particular, the WP will construct and demonstrate in pilot studies the use of a variational data assimilation system around a coupled atmosphere-ocean model. The components of the coupled model are the Planet Simulator (containing, as atmospheric component, PUMA, a light version of ECHAM) and the MIT's oceanic general circulation model (MITgcm). The system will be constructed to allow the expansion to a more complex atmosphere module at a later stage. At the same time the WP will lead to improved ocean syntheses used under WP 4.2 as initial conditions, by including new ocean data sets into the estimation procedure, including the ocean transports measured as part of this EU effort.
The specific tasks are:
• Coupling MITgcm to Planet Simulator and testing
• Running forecast experiments
• Preparation of observational atmospheric test data set
• Developing and testing variational data assimilation system around the Planet Simulator
• Planet Simulator assimilation
• Integration of coupled components
• Coupled MITgcm-Planet Simulator assimilation
It is anticipated that the outcome of the WP is a first pilot assimilation system that allows ocean data assimilation into coupled models and thereby will lead to much improved initialisation of coupled models, as it will be required for the next generation forecast models.
Lead: Detlef Stammer (UHAM)
Participants: UHAM, MPG-M, ECMWF, KNMI