Our goal of this project is to develop an urban scheme for land surface
model. Focusing on Houston, USA, we will use MODIS land surface
observations to better represent landscape and use skin temperature
measurements to validate model performance.
We will retrieve 1km, daily MODIS land surface observations of LAI,
albedo, emissivity, land cover, skin temperature and cloud/aerosol
observations from EOS data archive for Houston urban and suburban regions
from July to September 2000. Our data set will match the Texas Air Quality
Study field experiment conducted in August-September 2000 and will provide
a complete picture of land surface, atmosphere, and biosphere conditions
during that period of time.
Urban climate and aerosol transport are determined by the distribution of
building and roads, building density, and vegetation coverage, which are
in general referred to as "urban geometry". We will map the detailed urban
building, road, and canopy information from satellite visible observations
at 15m resolution. The 15m observation will then be scaled up into 1km for
model simulation. More importantly, we need to derive roughness length
based on building information for the use in land surface model.