With increasing computational resources, mesoscale numerical weather prediction (NWP) models can now be run at sub-kilometer grid spacing. The high resolution NWP forecasts can be used to initialize the transport and dispersion model for improved prediction of the urban meteorological conditions. In April 2008, an updated version of the 2-d Urban Canopy Model (UCM) coupled with WRFV3.0/Noah model was released. The improvement in this version of the UCM is the incorporation of a diurnal profile for the anthropogenic heat. In the present study, we have used the released version of WRFV3.0/UCM to perform a series of numerical experiments over the Houston region. Urban morphological gridded data sets developed under National Urban Database and Access Portal Tools (NUDAPT) are incorporated for these runs. One of the major challenges in using the WRF/UCM is the specification of the initial roof, wall and road temperatures since no routine observations are available. An urbanized version of the High Resolution Land Data Assimilation System (u-HRLDAS), which is developed at National Center for Atmospheric Research (NCAR), provides spin-up values of roof, wall and road temperatures. The roof, wall and road temperatures from u-HRLDAS are then used to initialize the WRF/UCM.

Using the latest version of the WRF/UCM, an u-HRLDAS initialized roof, wall, road temperature and NUDAPT datasets, a series of short term simulations are performed over the Houston region. The main objectives of these numerical experiments are to study (1) the sensitivity of different urban parameters specified through the urban parameter table, and (2) the impact of using the NUDAPT datasets on WRF simulations. For the evaluation of the numerical experiments the observational data obtained from the Texas Air Quality Study 2000 (TexAQS2000) field program are used. The results from these experiments would be discussed at the conference.