Final Report Summary - SANGOMA (Stochastic Assimilation for the Next Generation Ocean Model Applications)
Executive Summary:
SANGOMA provided a step towards adavanced data assimilation techniques for ocean model. For research version new assimilation techniques were developped whereas for operation uses, EnKF versions where devolpped in an open and common framework, easily uptaken by new data assimilation users. To test the developments, open benchmarks were defined and new data types analysed and included. In summary:
WP1 ensured the planned harmonization of data models and interfaces. This was not easy a priori and an "Keep it simple and short" approach was chosen. The objective has been achieved and, interestingly, the defined standards proven adequate after the fact, including for the advanced EWPF (Equivalent Weight Particle Filter) implementation.
WP2 provided a nice collection of data-assimilation tools and was elaborated with a complete set of examples. Particularly diagnostic tools with simple and advanced techniques are of wide interest. Also a common framework for a series of variants of EnKF (Ensemble Kalman Filter) was provided as a starting point for newcomers in data assimilation.
WP3 shares nice reviews and presentations of the EnKF variants in a coherent framework and terminology were synthesized. New theoretical methods were developed and tested, from which two (EWPF+Anamorphosis) were implemented into toolboxes with SANGOMA standards.
WP4 included benchmark definitions (taken up by other groups outside of SANGOMA) with new metrics, in particular probabilistic metrics for ensemble approaches.
In WP5 Realistic applications with new data types were implemented and analyzed.
Finally WP6 disseminated the results and achieved a strong scientific output.
\end{itemize}
Project Context and Objectives:
SANGOMA provided new developments in data assimilation to ensure that future operational systems make use of state-of-the-art data-assimilation and related analysis tools. We are a European network of expert teams in advanced data assimilation. In the project we extended existing modular data assimilation systems that have high flexibility in type of ocean model and assimilation method. Following specific design rules, new modules can be used in different modular systems. The systems allows for efficient operational testing of the latest data assimilation methods, and quick comparison of assimilation methods for operational use. Furthermore, we developped and implemented modules that objectively determine the impact of existing and new observation types.
This dedicated web portal provides access to validated products, including documented performances on a variety of test cases. Consolidated versions are made available to the science community and Marine Forecasting Centres with indications on best practice implementation. Workshops and summer schools on advanced assimilation methods and modular systems ensured fast and efficient training of next generation oceanographers, ensuring world-leading operational oceanographic products for costumers and decision makers.
Project Results:
See attached document.
Potential Impact:
The developments of SANGOMA will also serve costumers of MyOcean products, which is the first European project dedicated to the implementation of the GMES Marine Core Service for ocean monitoring and forecasting. For this purpose, we will concentrate on data-assimilation methods that deliver probabilistic information on the products. To this end, existing ensemble methods will be included and new methods that allow for nonlinear and non-Gaussian systems will be developed.
List of Websites:
http://www.data-assimilation.net/
SANGOMA provided a step towards adavanced data assimilation techniques for ocean model. For research version new assimilation techniques were developped whereas for operation uses, EnKF versions where devolpped in an open and common framework, easily uptaken by new data assimilation users. To test the developments, open benchmarks were defined and new data types analysed and included. In summary:
WP1 ensured the planned harmonization of data models and interfaces. This was not easy a priori and an "Keep it simple and short" approach was chosen. The objective has been achieved and, interestingly, the defined standards proven adequate after the fact, including for the advanced EWPF (Equivalent Weight Particle Filter) implementation.
WP2 provided a nice collection of data-assimilation tools and was elaborated with a complete set of examples. Particularly diagnostic tools with simple and advanced techniques are of wide interest. Also a common framework for a series of variants of EnKF (Ensemble Kalman Filter) was provided as a starting point for newcomers in data assimilation.
WP3 shares nice reviews and presentations of the EnKF variants in a coherent framework and terminology were synthesized. New theoretical methods were developed and tested, from which two (EWPF+Anamorphosis) were implemented into toolboxes with SANGOMA standards.
WP4 included benchmark definitions (taken up by other groups outside of SANGOMA) with new metrics, in particular probabilistic metrics for ensemble approaches.
In WP5 Realistic applications with new data types were implemented and analyzed.
Finally WP6 disseminated the results and achieved a strong scientific output.
\end{itemize}
Project Context and Objectives:
SANGOMA provided new developments in data assimilation to ensure that future operational systems make use of state-of-the-art data-assimilation and related analysis tools. We are a European network of expert teams in advanced data assimilation. In the project we extended existing modular data assimilation systems that have high flexibility in type of ocean model and assimilation method. Following specific design rules, new modules can be used in different modular systems. The systems allows for efficient operational testing of the latest data assimilation methods, and quick comparison of assimilation methods for operational use. Furthermore, we developped and implemented modules that objectively determine the impact of existing and new observation types.
This dedicated web portal provides access to validated products, including documented performances on a variety of test cases. Consolidated versions are made available to the science community and Marine Forecasting Centres with indications on best practice implementation. Workshops and summer schools on advanced assimilation methods and modular systems ensured fast and efficient training of next generation oceanographers, ensuring world-leading operational oceanographic products for costumers and decision makers.
Project Results:
See attached document.
Potential Impact:
The developments of SANGOMA will also serve costumers of MyOcean products, which is the first European project dedicated to the implementation of the GMES Marine Core Service for ocean monitoring and forecasting. For this purpose, we will concentrate on data-assimilation methods that deliver probabilistic information on the products. To this end, existing ensemble methods will be included and new methods that allow for nonlinear and non-Gaussian systems will be developed.
List of Websites:
http://www.data-assimilation.net/
