It’s a considerable step forward, as much for the applied sciences as for oceanography. Jean-Marie Beckers, a professor at the ULg, Hans Burchard and Richard Hofmeister have developed a numerical model which simulates the oceans’ hydrodynamics. This 3D model is capable of adapting according to the importance of the phenomena being studied.

When we think of an oceanographer we can very easily imagine an ersatz Commander Cousteau, flippers screwed to his feet, an oxygen tank strapped on his back, on the quest for new animal species in the dark waters of the oceans’ depths. Jean-Marie Beckers is for his part very far from matching the profile. He is nonetheless a professor in oceanography at the University of Liège’s department of astrophysics, geophysics and oceanography.

This trained engineer certainly has ambitions to explore the seas’ depths, but on the back of a mouse, and in front of a computer screen. Concerned with neither the marine flora nor its fauna, he instead has ambitions to pierce the mysteries of hydrodynamics, the circulation of the oceans’ waters, in taking into account the waters’ currents, salinity and temperature. ‘In reality,’ explains Jean-Marie Beckers, ‘there are three aspects connected to hydrodynamics. First of all there is observation, thus gathering and analysing data, enabling a more precise knowledge of the oceans, in particular concerning its present or past condition. At the other extreme there is the building up of numerical models aimed at describing the evolution of the oceans and to draw up predictions of oceanic circulation. These models are based on the laws of physics such as the preservation of mass and Newton’s law. Between the two we do what is called data assimilation. We incorporate the gathered data into the model to correct or improve the quality of the prediction.’

It is this modelling that we will in particular look into. Jean-Marie Beckers has just in effect jointly signed off on a publication bringing to light the development of an adaptive 3D numerical model. To better unearth the value of such a development, a little step backwards is required. Numerical modelling is the translation of the laws of physics, applied to the oceans, into a computer calculation algorithm. Discretisation is also brought into play: to study a unit of ocean horizontally (in 2D thus), the model’s designer will cut the unit into several cubes which resemble nothing less than the pixels in a photograph. At a given moment the water which leaves a cube through one of its sides enters the adjacent cube and so on. To each of these cubes a mass assessment as well as Newton’s law is applied. All of the grid’s cubes are connected between themselves and it is thus possible to calculate the evolution of the waters in the unit studied. As regards the allusion to photography it is easy to grasp that the more squares there are in the grid the more precise will the data be, which improves the quality of the model. This phenomenon is explained by the reduction of numerical mixing. ‘The only thing is that reducing the size of the squares considerably increases the number of calculations to be made,’ stresses the oceanographer, ‘which would require more time, better technology and thus a higher cost.’