Mesoscale Meteorological Modeling at Kilometer Scale Grid Meshes

We address a basic issue associated with mesoscale numerical weather prediction models used for weather forecasting and air quality assessments with fine [O(1km)] grid meshes. Such meshes are small enough to capture features of the larger convectively-induced secondary circulations (CISCs) such as evident in horizontal convective rolls and cellular vortices that may exist in the diurnal planetary boundary layer (PBL), but too large for the simulation of the turbulent cascade that regulates their amplitude and structure. Higher-resolution mesoscale models with current PBL parameterizations developed for larger-grid-mesh [O(10) km] simulations can produce realistic-looking CISC-like features occurring in the heated PBL but without sound theoretical foundation, and thus are inaccurate and misleading. We describe how the superadiabatic layer near the surface together with the increase of the horizontal resolution is responsible for the model-CISCs (M-CISCs). We also document their shortcomings, and explore some options for dealing with them. These include use of WRF-LES in the innermost domain to represent the role of turbulence in model (M-CISC) evolution; modification of PBL schemes so that they “take over” transport by M-CISCs, thus eliminating them; use of LES to develop parameterization schemes that represent the role of turbulence in M-CISC evolution, thus allowing simulation of robust M-CISCs; or doing nothing, namely allowing the M-CISCs to exist, but accounting for their shortcomings during interpretation. We also explore considerations for more operational applications when considering the role of land use variations including land-water and urban-nonurban contrasts.