Advancing airborne Doppler lidar wind measurements for atmospheric boundary layer research

There is a need for improvement of observational capabilities with respect to wind measurements in the atmospheric boundary layer (BL). This work is dedicated to the commissioning of a new airborne Doppler lidar (ADL) system, which provides wind measurements inside the turbulent BL at highest resolution up to date. To assess the potential of such measurements and address new challenges presented by them, a two-step approach based on virtual observations and real-world measurements is employed. As a first part of this work, a novel large eddy simulation (LES)-based airborne Doppler lidar simulator (ADLS) is developed. In the ADLS, simulated measurements are conducted based on LES wind fields, which provides the advantage of knowing the 3-D wind field along the lidar measurement curtain. In a first step, the ADLS is used to assess the error in ADL retrieved wind profiles, due to the violation of the flow homogeneity assumption postulated in the retrieval, for a common system setup and retrieval strategy. In a second step, the ADLS is used to identify preferable system setups and to quantify retrieval strategy limits beyond which ADL wind profiling becomes unreliable. Additionally, the ADLS is used to validate a proposed data-driven uncertainty estimation method, offering new possibilities for wind profile accuracy assessment. In the second part of this work, real-world measurements obtained on four research flights with a new prototype ADL system are presented. In data processing, the motion correction accuracy, time synchronization, beam alignment and lidar signal quality control is evaluated. It is shown that the measured ground return velocities allow for a flight state dependent assessment of error in the inertial navigation system, which is used for motion correction. A refined ground return based motion correction procedure is demonstrated to be beneficial for measurement accuracy. The results show that reliable measurements are possible even when the aircraft is flying inside the turbulent BL. Thus, in combination with the ADLS results, the accuracy of the retrieved wind profiles is assessed. Further, it is shown that the vertical wind mean and variance, and their associated uncertainties, can be quantified from ADL measurements by taking into account the fixed-beam geometry, lidar pulse averaging effects, as well as random and systematic uncertainties due to the finite measurement lengths. Finally, the analysis of three application studies demonstrates that the prototype system provides substantial new insight into flow and turbulence characteristics of the BL