Errors for space-based Doppler lidar wind measurements: Definition, performance, and verification

Verification of space-based wind measurements will be difficult because of the random variations of the atmospheric velocity field over the measurement volume. The definition of accuracy requires a definition of "truth.'' For this work, truth is defined as the spatial average of the random velocity over the measurement volume. The bias of space-based velocity measurements is dominated by the pointing errors of the lidar beam. The bias and random errors of Doppler lidar measurements produced by the estimation algorithm and the random atmospheric parameters are determined with computer simulations for both the sampling errors of the lidar scanning pattern and the estimation error of the radial velocity measurements. The results are compared with the error produced by ideal rawinsonde measurements and aircraft winds. A simple method to verify the random error of Doppler lidar radial velocity measurements using only multiple-shot lidar data (no in situ data are required) is compared with results from computer simulation. The verification of space-based measurements using the correlation between rawinsonde measurements and aircraft winds is also investigated for typical tropospheric conditions. For coherent Doppler lidar, the random errors are dominated by the sampling or representative errors produced by the scanning geometry of the lidar beam. The instrumental random error is typically less than the sampling error.