Estimation of wind velocity and backscatter signal intensity from Doppler lidar returns

Abstract

Doppler lidar offers a method of remotely measuring wind speeds in optically clear air. A laser source directs a pulse of narrowband optical radiation into the atmosphere. As the pulse propagates, it irradiates small aerosol particles, which scatter radiation back toward the lidar system. Because the aerosol particles are small enough to be borne by the wind, the frequency of the scattered radiation is Doppler shifted as a result of motion toward or away from the lidar. Measurement of this Doppler shift provides an estimate of the radial wind component, while computation of the power in the backscattered radiation gives insight into the atmospheric turbidity, presence of aerosol layers or clouds, and attenuation. Pulsed Doppler lidar systems have been used in a number of meteorological applications. Deployment of a Doppler lidar on an Earth-orbiting satellite has been proposed to remotely measure tropospheric winds on a global scale. In a large number of these applications, lidar system performance, impact, and maximum range is limited by weak backscattered signals present at the lidar receiver. Because potential methods of improving signal-to-noise ratio can be technically infeasible and/or very expensive, optimizing signal processing of the lidar return to improve performance at very low signal levels has received significant attention. In the remainder of this paper, we describe techniques for processing Doppler lidar returns from weak and fluctuating signals, and discuss system design tradeoffs to obtain maximum performance.