Urban Aerodynamic Roughness Length Mapping Using Multitemporal SAR Data

Abstract

Aerodynamic roughness is very important to urban meteorological and climate studies. Radar remote sensing is considered to be an effective means for aerodynamic roughness retrieval because radar backscattering is sensitive to the surface roughness and geometric structure of a given target. In this paper, a methodology for aerodynamic roughness length estimation using SAR data in urban areas is introduced. The scale and orientation characteristics of backscattering of various targets in urban areas were firstly extracted and analyzed, which showed great potential of SAR data for urban roughness elements characterization. Then the ground truth aerodynamic roughness was calculated from wind gradient data acquired by the meteorological tower using fitting and iterative method. And then the optimal dimension of the upwind sector for the aerodynamic roughness calculation was determined through a correlation analysis between backscattering extracted from SAR data at various upwind sector areas and the aerodynamic roughness calculated from the meteorological tower data. Finally a quantitative relationship was set up to retrieve the aerodynamic roughness length from SAR data. Experiments based on ALOS PALSAR and COSMO-SkyMed data from 2006 to 2011 prove that the proposed methodology can provide accurate roughness length estimations for the spatial and temporal analysis of urban surface.