Estimating spatial velocity statistics with coherent Doppler lidar

The spatial statistics of a simulated turbulent velocity field are estimated using radial velocity estimates from simulated coherent Doppler lidar data. The structure functions from the radial velocity estimates are processed to estimate the energy dissipation rate ∉ and the integral length scale L_i, assuming a theoretical model for isotropic wind fields. The performance of the estimates are described by their bias, standard deviation, and percentiles. The estimates of ∉^2/3 are generally unbiased and robust. The distribution of the estimates of L_i are highly skewed; however, the median of the distribution is generally unbiased. The effects of the spatial averaging by the atmospheric movement transverse to the lidar beam during the dwell time of each radial velocity estimate are determined, as well as the error scaling as a function of the dimensions of the total measurement region. Accurate estimates of L_i require very large measurement domains in order to observe a large number of independent samples of the spatial scales that define L_i.