CMEMS-Based Coastal Analyses: Conditioning, Coupling and Limits for Applications
Agustín Sánchez‐ArcillaJoanna StanevaLuigi CavaleriMerete BadgerJean‐Raymond BidlotJacob Tornfeldt SørensenLars Boye HansenAdrien MartinAndy SaulterManuel EspinoMario Marcello MigliettaMarc MestresDavide BonaldoPaolo PezzuttoJohannes Schulz‐StellenflethAnne WieseXiaoli Guo LarsénSandro CarnielRodolfo BolañosSaleh AbdallaAlessandro Tiesi
25
Citation
79
Reference
10
Related Paper
Citation Trend
Abstract:
Recent advances in numerical modeling, satellite data, and coastal processes, together with the rapid evolution of CMEMS products and the increasing pressures on coastal zones, suggest the timeliness of extending such products toward the coast. The CEASELESS EU H2020 project combines Sentinel and in-situ data with high-resolution models to predict coastal hydrodynamics at a variety of scales, according to stakeholder requirements. These predictions explicitly introduce land discharges into coastal oceanography, addressing local conditioning, assimilation memory and anisotropic error metrics taking into account the limited size of coastal domains. This article presents and discusses the advances achieved by CEASELESS in exploring the performance of coastal models, considering model resolution and domain scales, and assessing error generation and propagation. The project has also evaluated how underlying model uncertainties can be treated to comply with stakeholder requirements for a variety of applications, from storm-induced risks to aquaculture, from renewable energy to water quality. This has led to the refinement of a set of demonstrative applications, supported by a software environment able to provide met-ocean data on demand. The article ends with some remarks on the scientific, technical and application limits for CMEMS-based coastal products and how these products may be used to drive the extension of CMEMS toward the coast, promoting a wider uptake of CMEMS-based predictions.Data assimilation applications with large-scale numerical models
exhibit extreme requirements on computational resources. Good
scalability of the assimilation system is necessary to make these
applications feasible. Sequential data assimilation methods based on
ensemble forecasts, like ensemble-based Kalman filters, provide such
good scalability, because the forecast of each ensemble member can be
performed independently. However, this parallelism has to be combined
with the parallelization of both the numerical model and the data
assimilation algorithm. In order to simplify the implementation of
scalable data assimilation systems based on existing numerical models,
the Parallel Data Assimilation Framework PDAF (http://pdaf.awi.de) has
been developed. PDAF provides support for implementing a data
assimilation system with parallel ensemble forecasts and parallel
numerical models. Further, it includes several optimized parallel
filter algorithms, like the Ensemble Transform Kalman Filter.
We will discuss the philosophy behind PDAF as well as features and
scalability of data assimilation systems based on PDAF on the example
of data assimilation with the finite element ocean model FEOM.
Ensemble forecasting
Ensemble Learning
Assimilation (phonology)
Cite
Citations (0)
Data assimilation applications with large-scale numerical models
exhibit extreme requirements on computational resources. Good
scalability of the assimilation system is necessary to make these
applications feasible. Sequential data assimilation methods based on
ensemble forecasts, like ensemble-based Kalman filters, provide such
good scalability, because the forecast of each ensemble member can be
performed independently. However, this parallelism has to be combined
with the parallelization of both the numerical model and the data
assimilation algorithm. In order to simplify the implementation of
scalable data assimilation systems based on existing numerical models,
the Parallel Data Assimilation Framework PDAF (http://pdaf.awi.de) has
been developed. PDAF provides support for implementing a data
assimilation system with parallel ensemble forecasts and parallel
numerical models. Further, it includes several optimized parallel
filter algorithms, like the Ensemble Transform Kalman Filter.
We will discuss the philosophy behind PDAF as well as features and
scalability of data assimilation systems based on PDAF on the example
of data assimilation with the finite element ocean model FEOM.
Assimilation (phonology)
Ensemble forecasting
Cite
Citations (0)
Data assimilation applications with high-dimensional numerical modelsshow extreme requirements on computational resources. Thus, goodscalability of the assimilation system is necessary to make theseapplications feasible. Sequential data assimilation methods based onensemble forecasts, like ensemble-based Kalman filters, provide suchgood scalability, because the forecast of each ensemble member can beperformed independently. However, this parallelism has to be combinedwith the parallelization of both the numerical model and the data assimilation algorithm. In order to simplify the implementation ofscalable data assimilation systems based on existing numerical models,the Parallel Data Assimilation Framework PDAF has been developed. Itprovides support for parallel ensemble forecasts and parallelnumerical models. Further, it includes several optimized parallel filteralgorithms, like the ensemble transform Kalman filter. We will discussthe features and scalability of data assimilation systems based onPDAF on the example of data assimilation with the finite element oceanmodel FEOM.
Assimilation (phonology)
Cite
Citations (0)