Study Of Aerodynamic Roughness Length Andthermal Roughness Length On Different Underlying Surfaces

The land-atmosphere interaction is not only an important part of the climate system, but also the foundation of studying the earth system and global climate change. It provides a more reliable basis for the improvement of regional energy balance estimate and atmospheric numerical model, in-depth understanding of land-atmosphere interaction by describing the key land surface parameters （z0m, z0h） more reasonably and more aptly. In this paper, based on the data of Eddy Covariance （EC） system and Automatic Weather Station （AWS） collected at Guantao, A’Rou sites in2010and Beijing325m tower in2007, the value of z0m and zoh are estimated. And then the characteristics of spatial and temporal change are analyzed with these estimation values, which are compared with the ones modeling by different parameterization schemes. The result shows that（1） Yang et al.（2003） is the best method of all estimation ones chose here and its calculation results reveals seasonal variation of z0m in both crop and grass surfaces accurately. Contrast to A’Rou site, the directional characteristics of z0m is obvious in Guantao site. The value of aerodynamic roughness length changes with wind direction due to the morphological character difference of the roughness elements within corresponding source areas on various wind direction. In addition, the z0m is essentially related to characteristics of the roughness elements, which showing a linear relationship with the vegetation height and a parabolic relationship with Leaf Area Index （LAI）. Meanwhile, different parameterization method has its own limitations on various land types, in which the results using M_zom4and M_zoml is most close to the estimation ones at Guantao and A’Rou sites, respectively.（2） The values of kB-1have a obvious range on different underlying surface, which has a diurnal variation and also a good correlation with surface-air temperature difference, except maize and maize/bare soil surface. On uniform vegetation, the change of kB-1exhibits a normal distribution pattern. The sensible heat flux error is about2%with a median or mean value of kB-1, so that kB-1can set at a fixed value on this surface. On others, kB-1is a changing value and need model modeling. Comparing sensible heat flux using seven parameterization schemes with the observed values, results of M₁989a and M₁998are most close to the observe value on bare soil while on mixed terrain surface, M₁958, M₁963and M₂007b are the best ones compared with observe value. M₁982and M₂002both have limitation on all surfaces.（3） On Beijing city underlying surface, z0m has obvious directivity,1.78m in the prevailing wind direction from90degree to180degree. kB-1has obvious variation range, and has a certain pattern of frequency distribution. Comparing with underlying surface of suburb （farmland and grassland）, statistical values of kB-1is significantly higher.