Improvement of VAD wind profile from Doppler lidar using accumulated spectra

Pulsed coherent Doppler lidar (CDL) is a remote sensing instrument for measuring the wind field and turbulence statistics of boundary layer processes. The CDL-retrieved vertical wind profile is commonly obtained using the velocity-azimuth-display (VAD) algorithm under the assumption of homogeneous wind field. The quality of the retrieved wind field highly depends on the accuracy of the radial velocity, which is influenced by the signal-to noise ratio (SNR). The decrease of the SNR with increasing range often limits the height of reliable retrieved wind profile, sometimes resulting in not fully observing boundary layer processes. The SNR can be enhanced by increasing the number of pulses in incoherent integration, and/or the number of samples along the range direction. However, these approaches come at the expense of temporal and/or spatial resolutions, respectively. In the previous studies, the depth of vertical wind profile from CDL can be improved using the maximum of the function of accumulated spectra (MFAS) technique. Without estimating radial velocities at each azimuth angle, the horizontal wind is estimated by accumulating Doppler velocity bins in each spectrum. As a result, the wind field can be estimated at lower SNR comparing to the traditional VAD method, leading to the extended height of the retrieved wind profile. However, the MFAS-derived wind vector has not been validated with real measurement.

In the study, the MFAS-derived horizontal wind vector is first validated by the sounding observations. Both high correlation coefficient (>0.99) for zonal and meridional wind shows the MFAS-derived wind vector will not degrade the results where the VAD is valid. The root-mean-square deviation between MFAS-derived wind speed and the sounding measurement is less than 1.5 m/s in the region where VAD does not provide reliable result. With the enhanced performances, the averaged increase of the effective height is up to approximately 1 km. As a result, the MFAS-derived wind vector extends the effective range of existing CDL systems without sacrificing spatiotemporal resolution. The frequency of measuring entirety of the boundary layer using MFAS compared to only traditional VAD will be quantified.