Data-driven numerical modeling in geodynamics methods and applications / Численное моделирование в геодинамике: методы и приложения

Dynamic processes in the Earth’s interior and on its surface can be described by geodynamic models. These models can be presented by a mathematical problem comprising a set of partial differential equations with relevant conditions at the model boundary and at the initial time. The mathematical problem can be then solved numerically to obtain future states of the model. Meanwhile the initial conditions in the geological past or some boundary conditions at the present are unknown, and the question of how to “find” the conditions with a sufficient accuracy attracts attention in the field of geodynamics. One of the mathematical approaches is data assimilation or the use of available data to reconstruct the initial state in the past or boundary conditions and then to model numerically the dynamics of the Earth starting from the reconstructed conditions. <...>