Does Land-Surface Heterogeneity Affect the Evolution of Hurricanes after their Landfall in the Eastern US - A Numerical Study with the Ocean-Land-Atmosphere Model (OLAM)

The Ocean-Land-Atmosphere Model (OLAM) has been developed to extend the capabilities of the Regional Atmospheric Modeling System (RAMS) to a global modeling framework. OLAM is a new model with regard to its dynamic core, grid configuration, memory structure, and numerical solution technique. Instead of the Boussinesq approximation used in RAMS, OLAM solves the full compressible Navier-Stokes equations in conservation form using finite-volume numerical operators that conserve mass, momentum, and energy to machine precision. In place of RAMS' structured multiple nested grids and hexahedral grid cells on a polar stereographic projection, OLAM uses a single unstructured grid and pentahedral (prism) grid cells (with a triangular footprint) which conform to the sphere without a coordinate transformation. OLAM's grid topology enables local mesh refinement to any degree without the need for special grid nesting algorithms; all communication between regions of different resolution is accomplished seamlessly by flux-conservative advective and diffusive transport. OLAM represents topography using a form of the volume-fraction or shaved grid cell method in which model levels are strictly horizontal, rather than terrain-following, and therefore intersect topography. Grid cell face areas, which explicitly appear in the finite volume equations and are pre- computed and stored, are reduced in proportion to any blockage by topography, thereby correctly regulating inter-cell transport and preventing advective flux normal to the ground surface. Apart from its dynamic core and grid configuration, OLAM bears a strong resemblance to RAMS. Both models share the same physical parameterizations for microphysics, land and vegetation water and energy balances, radiative transfer, and sub-grid cumulus convection. Model coding structure, I/O file formats, and methods of compiling, initializing, and executing the models are very similar or identical. OLAM is particularly well designed to simulate weather events such as hurricanes, which require a high- resolution grid in the region of the hurricane to properly simulate the eye of the storm but also need to interact with the large-scale dynamics, which they significantly affect and are affected by through complex interactions. Here, we investigate with OLAM the impact of land-surface heterogeneity on the evolution of hurricanes after their landfall in the eastern US. We use Hurricane Frances as a case study.