Abstract
The analysis of airborne Doppler lidar data taken by NASA near the top of the planetary boundary layer in the central valley of California is presented. These data include downward wan angles that intercept the ground. The maximum errors in the radial speeds, based on the apparent speed of the ground strikes, was found to have a magnitude of less than 2.0 m s−1. These errors appear to have a slowly varying systematic part and a random part. The tidal radial speeds were corrected for running averages of these errors resulting in a random residual error of less than 0.3 m s−1 in magnitude. Data are considered usable for those range bins in which the return signal amplitude exceeds that of the average shot noise by two times the rms variation in the noise. These data were analyzed to produce wind vectors on a geographic grid divided into 1 kilometer square cells. The resulting wind field appears to be very realistic and contains features similar to those apparent in ground level wind observations. The vertical shear in the sea breeze is well resolved as is its apparent containment below well-defined inversions. While the analysis was labor intensive and cumbersome at this early stage of system development, the final product is a heretofore unattainable high resolution depiction of air flow over a broad region. The analyzed wind fields appear to be very realistic and easy to interpret. The technique is accurate in the sense that wheel the flow is slowly varying relative to the time sale of the measurement technique, the accuracy of the system is comparable to more standard techniques but with much greater spatial resolution than possible with other techniques. Where the flow is nonstationary, the recovered wind fields show this as well, i.e., the variability of the wind is correctly indicated. The success of this experiment is due to the high precision of the modified inertial navigation unit used and the use of ground strikes to provide absolute verification of total system performance.