Estimates of effective aerodynamic roughness length over mountainous areas of the Tibetan Plateau

Wind profile data were measured by a wind profiler at QOMS (The Qomolangma Station for Atmospheric Environmental Observation and Research, Chinese Academy of Sciences) and, in separate experiments, GPS radiosondes were used at Shiquanhe Station and Litang Station on the Tibetan Plateau (TP). All three stations are located in the most rugged areas of the TP. QOMS is surrounded by the Himalaya, Shiquanhe Station by the Transhimalaya and Litang station is near the Hengduan Mountains. Using observational wind profile data, effective aerodynamic roughness lengthz 0 m effand zero‐plane displacement height d 0 are determined using the neutral logarithmic wind profile law. The results show that the values ofz 0 m effderived from the wind profiles can be considerably larger than the small‐scale aerodynamic roughness lengths of the land surface around the three stations. Subsequently, several parametrization schemes which use land surface characteristics to estimatez 0 m effand d 0 , such as roughness obstacle height and density, were assessed. The result indicates that of all the methods available, that proposed by Grant and Mason, where the drag coefficient D = 0.5, gives the best estimate ofz 0 m eff . The interim European Centre for Medium‐range Weather Forecasts (ECMWF) Re‐Analysis (ERA‐Interim) dataset underestimates effective aerodynamic roughness length in mountainous areas of the TP due to its use of a smaller drag coefficient of D = 0.4. For estimating d 0 , the method of Kutzbach performs well when the density of roughness obstacles ( λ ) is low, whereas the method of Raupach et al. gives a more reliable estimate when λ is high. Although this analysis has some limitations, it can feasibly account for form drag being exerted by an unresolved topography in the mountainous areas of the TP.