Four dimensional data assimilation of dual doppler lidar observations of the urban boundary layer

A better understanding of transport processes in the atmospheric boundary layer (ABL) is essential for emergency action in cases of chemical and biological agent dispersion. However, our current knowledge of ABL processes is limited, due in large part to frequently changing large-scale forcings and the lack of measurements of nearly any atmospheric variable at higher altitudes. To gain new insights into the ABL transport processes and the dispersion of contaminants in cities to address some concerns of homeland security, an atmospheric dispersion study was held in Oklahoma City, Oklahoma in 2003. Two Doppler Light-Detection-And-Ranging (lidar) systems were deployed to collect hours of radial velocity and aerosol concentration data in both the daytime and nighttime boundary layers. These lidar data permit the study of ABL flow characteristics and the atmospheric dispersion of particles in an urban setting.;The four dimensional variational data assimilation method (4DVAR) is adopted to merge limited lidar observations with a computational fluid dynamics model to derive detailed three-dimensional (3D) wind and temperature data. The accuracy of the 4DVAR method is assessed by comparing the single lidar retrieval with both the second lidar measurement and the dual lidar retrieval. The single lidar 4DVAR has proven to retrieve accurate flow fields even without the cross-beam information. The turbulent flow characteristics of the retrieved 3D wind and temperature field are studied. A number of coherent structures have been identified by the proper orthogonal decomposition method. The representation of instantaneous snapshots by high-ranking eigenmodes is examined by the reconstruction of reduced-order flow fields. The Lagrangian particle dispersion model has been successfully applied to predict turbulent dispersion in the convective and stable urban ABL. The effects of different source locations and heights have been examined. The vertical mixing is slower in the stable ABL due to table stratification.