A Fast and Generic Method to Identify Parameters in Complex and Embedded Geophysical Models: The Example of Turbulent Mixing in the Ocean

Geophysical systems are generally described by nonlinear mathematical models. These models often involve partial differential equations like the Navier-Stokes equations, and embed sub-models describing various processes, such as sub-grid turbulent mixing or biological processes (e.g., growth and interactions of living organisms). The resulting models are costly to develop and to run, bringing together scientists from different disciplines. Moreover, the resulting model predictions remain sensitive to various forms of uncertainty (