Internal Research Project I-09

January 2024 - June 2024

Countless systems in the environmental, technical and biological world can be described as coupled freeflow and porous media flow systems. These systems are often difficult to simulate due to the complex dynamics at their coupled interface. Although the coupled REV scale models developed and used in research at the moment are very powerful and useful tools, they often neglect important aspects of the sub-scale systems. The SFB 1313 works to develop modelling techniques that can describe these coupled systems across a few scales of analysis by comparing sub-REV models with REV scale results. Another method for improving the coupled REV scale models would be to extend the current model scope with additional parameterized terms. These parameterized terms can then be used to include the neglected physics seen at the pore scale with averaged representations. Paramount to  the sucess of this method will be accurate parameterizations, and an understand of what factors effect these parameterizations.

Through existing projects, a framework for the data-driven parameterization of these terms has been developed. First, pore-scale simulations on periodic random pore geometries can be developed, which can be averaged and evaluated to isolate specific phenomna. Second, optimal parameters fitting to an REV scale concept can be found to match the missing isolated pore-scale phenomena. Third, the development of these optimal parameters can be replaced by a data-driven model, connecting pore-scale geometries to the REV-scale parameters. Lastly, these REV scale parameters can be connected to other common metrics used to describe porous media, in order to develop a proper understanding of what characteristics of porous medium will determine these optimal parameters.

This work has begun in a separate project, where investigations to dispersive transport in a porous medium have been developed. This can simply be extended to the scope of the PA-A in SFB 1313 by applying the same framework to questions of momentum evaluations in a porous medium (non-darcy regiemes), and tangential velocity conditions at an interface (Coupling conditions). Additionally, evaluations of roughness in the near
surface free-flow regeime could be developed and parameterized the same way. With these parameterizations in place, more effective evaluations of target applications, like soil-water evaporation, will be possible.