Sea Surface Temperature computed by the Mercator global model
(Credits Mercator Ocean)
Satellite observations provide a unique opportunity to monitor changes in the ocean in real time, accurately, on a global scale, and with high resolution. As only the properties of the sea surface can be observed from Space, data assimilation systems are needed to improve the consistency between satellite data and model simulations, to dynamically extrapolate and interpolate measurements scattered in space/time, and to better exploit the results of observations.
Ocean models are based on the laws of physics applied to a fluid (Newton’s laws, along with thermodynamics). Data assimilation is a procedure that combines actual observations with models. This combination aims to better estimate and describe the state of a dynamic system — the ocean. Estimates of this dynamic system are improved by correcting model errors with the observations on the one hand, and by synthesizing observations with the model on the other.
The vast majority of assimilation methods used in oceanography and meteorology are based on minimising the squared difference between the observations and their modelled equivalents. Several different approaches are used to resolve this problem: they are divided into so-called sequential and variational methods (such as 3D-Var, 4D-Var). Sequential methods, which are more basic but also more robust, enable the density of a water column to be modified in the model in order to correspond to the satellite-observed sea level height. These calculations are performed sequentially and the results are compared with observations at regular intervals, thus allowing the model to be corrected. Other more advanced techniques are costlier and more difficult to adjust due to their complexity. They include the variational method which aims to correct the initial conditions of the numerical model in order to obtain results that best fit the observations made throughout the period studied. At any point in time, therefore, the correction takes into account past and future observations. This technique is useful for controlling large-scale changes in ocean circulation and monitoring phenomena such as equatorial waves.
|Sea level anomalies on 8 January 2003 from the medium-resolution Mercator model (1/3°), with assimilation of in situSST measurements and (left) Jason-1 data only, (right) all available altimetry satellite data (Jason-1, ERS-2 and GFO). More detail can be seen in the right-hand figure.
(Credits Mercator Ocean)
Ocean models have now improved to the point where they are used to simulate and study the actual circulation of the ocean.
– Fukumori, I., Data assimilation by models, Satellite altimetry and Earth sciences, L.L. Fu and A. Cazenave Ed., Academic Press, 2001
– Ocean Weather Forecasting, E. Chassignet and J. Verron Ed.,Springer, 2006