Imposing land-surface fluxes at an immersed boundary for improved simulations of atmospheric flow over complex terrain

Boundary layer flows are greatly complicated by the presence of complex terrain which redirects mean flow and alters the structure of turbulence. Surface fluxes of heat and moisture provide additional forcing which induce secondary flows, or can dominate flow dynamics in cases with weak mean flows. Mesoscale models are increasingly being used for numerical simulations of boundary layer flows over complex terrain. At high spatial resolutions, particularly those needed to explicitly resolve complex terrain, mesoscale models become limited by the use of terrain-following coordinates which introduce numerical errors over steep gradients. The immersed boundary method alleviates errors associated with the coordinate transformation by allowing the terrain to be represented as a surface which arbitrarily passes through a Cartesian grid. Boundary conditions must then be imposed on the immersed surface for velocity and scalar surface fluxes.

This presentation describes coupling a land-surface model to an immersed boundary method in the Weather Research and Forecasting (WRF) model, allowing scalar fluxes at the immersed interface of explicitly resolved complex terrain. A new algorithm has been developed which allows scalar surface fluxes to be imposed on the flow solution at an immersed boundary. Previous algorithms impose Dirichlet boundary conditions at the immersed surface, but have not adequately addressed flux boundary conditions. With this extension of the immersed boundary method, land-surface models can be coupled to the immersed boundary to provide realistic surface forcing. Validation is provided in the context of idealized valley simulations with specified surface fluxes using the WRF code. Applicability to real terrain is illustrated by coupling a land-surface model to a two-dimensional immersed boundary representing the Owens Valley in California.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.