Abstract
The meteorological field over the complex terrain is important in simulating environmental air quality. In order to estimate precisely the meteorological field, we need to make good initial and boundary conditions for numerical simulation. To investigate the influence of different resolution of the bottom boundary condition on atmospheric components, we carried out numerical experiments. The experiments was performed with observation data to analyze the outputs of mesoscale model for varying bottom boundary conditions. The difference of topography for bottom boundary condition was USGS DEM with a grid distance of 1.1 km and the Ministry of Environment DEM (MDEM) with a grid distance of 90 m. These MDEM data were inserted into MM5 TERRAIN module through its modification. When comparing the temperatures, wind speeds from the numerical experiments with different bottom boundary conditions, meteorological variables with MDEM data agreed well with observation than those with USGS DEM, and especially improved to estimate in the portion of thermal area and lower wind speed (< 0.5 $ms^{-1}$) at 0000, 0600, 1200, and 1500 LST. In surface data and their numerical interpolation for improving the interpretation of meteorological components, objective analysis scheme should perform a smooth interpolation, detect and remove the bad data and carry out internal consistency analysis. For objective analysis technique which related to data reliability and error suppression, we carried out two quality control methods. First, as a site quality control (Q.C.1), removing the effect of terrain and buildings, has more effects on the wind speed than the temperature field. Second, as a quality control for availability of observed data (Q.C.2), temperature field has more improved than the wind speeds.