Aerodynamic roughness length estimation from very high-resolution imaging LIDAR observations over the Heihe basin in China

摘要

Roughness length of land surfaces is an essential variable for theparameterisation of momentum and heat exchanges. The growing interest in theestimation of the surface turbulent flux parameterisation from passiveremote sensing leads to an increasing development of models, and the commonuse of simple semi-empirical formulations to estimate surface roughness.Over complex surface land cover, these approaches would benefit from thecombined use of passive remote sensing and land surface structuremeasurements from Light Detection And Ranging (LIDAR) techniques. Followingearly studies based on LIDAR profile data, this paper explores the use ofimaging LIDAR measurements for the estimation of the aerodynamic roughnesslength over a heterogeneous landscape of the Heihe river basin, a typicalinland river basin in the northwest of China. The point cloud obtained frommultiple flight passes over an irrigated farmland area were used to separatethe land surface topography and the vegetation canopy into a DigitalElevation Model (DEM) and a Digital Surface Model (DSM) respectively. Thesetwo models were then incorporated in two approaches: (i) a strictlygeometrical approach based on the calculation of the plan surface densityand the frontal surface density to derive a geometrical surface roughness;(ii) a more aerodynamic approach where both the DEM and DSM are introduced ina Computational Fluid Dynamics model (CFD). The inversion of the resulting3-D wind field leads to a fine representation of the aerodynamic surfaceroughness. Examples of the use of these three approaches are presented forvarious wind directions together with a cross-comparison of results onheterogeneous land cover and complex roughness element structures.