Coupling Models and Data: which possibilities for remotely-sensed images?

The international scientific context is pointing out the role that can be played by models and observation systems for the evaluation and forecasting of the risks related to environmental problems. In this context, coupling data and models becomes a scientific challenge regarding the following considerations: A numerical model without observation data is not really interesting; On the other hand models lack the necessary data for forecasting, since they require an accurate initial condition for expecting good results and are making use of parameters which can not be obtained by measurement, for example sub-mesh parameterization. There is thus a need for an optimal use of all the available information, in order to constraint as best as possible models with their input data and parameters. Data assimilation (variational methods or Kalman filters) is a solution for combining different types of information: data and models. However structured spatial information of images are not often used in this context. If they are, this is in a qualitative way and not with accurate quantitative information. This paper is then considering the problem of Image Data Assimilation by focusing on satellite acquisition and on oceanographic applications. It is evaluating the different quantitative measures that could be obtained on these data for assimilation within these models.