Introducing Fractal Dimension into the Coupled WRF-Urban Model to Represent the Morphological Characteristics of Urban Canopy

With an ever-increasing portion of the population living in cities, land-surface characteristics have substantially changed in urbanized areas, which inevitably affect local weather and climate. To represent effects of urban expansion on the local weather, urbanization schemes have been introduced into atmospheric weather and climate models such as the coupled Weather Research and Forecasting (WRF)-Urban modeling system. In the WRF-Urban model, the percentage of impervious surface is used to characterize urban land-use condition, but using this percentage alone does not consider the morphological characteristics of the urban canopy. Therefore weather prediction is limited in their ability to perform accurate assessments in urban area. To better capture influences of the cities on wind and temperature in the urban boundary layer, we develop fractal dimension as a new approach to represent the three-dimensional characteristics of urban canopies. Fractal dimension combined with the percentage of impervious surface are employed to characterize the heterogeneity of the urban surface, so the description of the urban land use is improved. In this study we add a new parameter in WRF/SLUCM model, which is derived from fractal dimension and assigned as bu_index in WRF pre-processor. So the urban parameters related to three urban land-use classes are derived from both percentage of impervious surface and fractal dimension, instead of from the percentage of impervious surface in current WRF/SLUCM. We conduct a test case over the Beijing metropolis, and results illustrate that the simulated distribution of 10-m wind is more realistically associate with the building distribution. With the new approach, the simulated wind between the second and fourth ring roads in Beijing, due to higher buildings, is smaller than surrounding area. The 2-m temperature is lower than the original simulation. Thus improving the description of the 3-D characteristics of urban canopies could provide more accurate forecasts for urban regions.