Incorporating urban systems into global climate models: part 2, parameterization

Trish Jackson, Pei-Ling Lin, John Bauer, and Johannes J. Feddema
University of Kansas

An important aspect of incorporating an urban model into a GCM requires the accurate representation of urban characteristics through space and time. The nature of urban areas changes significantly based on population, cultural, socioeconomic and physical factors. The newly developed Urban Canyon Model for the National Center for Atmospheric Research (NCAR) Community Climate System Model (CCSM), requires information on building heights, building material properties (e.g. conductivity, and heat capacity), plan area vegetation and plan area building fractions, height to width ratios and information on the aerial extent of different urban sub-regions (e.g. central business district and industrial areas, medium to high density housing areas and suburban areas). Large cities throughout the world have widely varying characteristics, especially in terms of population density and urban sprawl. To define boundaries between urban and suburban areas we combined information from at multiple global land cover datasets as well as the Landscan population dataset, and continuous vegetation fraction information. Five land cover datasets were un-projected into a Geographic grid. Twenty-seven regions were defined that were believed to have similar urban characteristics. From the land cover datasets, we used urban areas as a mask to capture population data for these densely populated regions. Using the results from overlays between the datasets we evaluated the links between population density and measured urban characteristics.