Optimization of the Cross-Correlation Algorithm for Two-Component Wind Field Estimation from Single Aerosol Lidar Data and Comparison with Doppler Lidar

Abstract

Numerical and field experiments were conducted to test an optimized cross-correlation algorithm (CCA) for the remote sensing of two-component wind vectors from horizontally scanning elastic backscatter lidar data. Each vector is the result of applying the algorithm to a square and contiguous subset of pixels (an interrogation window) in the lidar scan area. Synthetic aerosol distributions and flow fields were used to investigate the accuracy and precision of the technique. Results indicate that in neutral static stability, when the mean flow direction over the interrogation window is relatively uniform, the random error of the estimates increases as the mean wind speed and turbulence intensity increases. In convective conditions, larger errors may occur as a result of the cellular nature of convection and the dramatic changes in wind direction that may span the interrogation window. Synthetic fields were also used to determine the significance of various image processing and numerical steps used in the CCA. Results show that an iterative approach that dynamically reduces the block size provides the largest performance gains. Finally, data from a field experiment conducted in 2013 in Chico, California, are presented. Comparisons with Doppler lidar data indicate excellent agreement for the 10-min mean wind velocity computed over a set of 150 h: the root-mean-square deviations (and slopes) for the u and υ components are 0.36 m s−1 (0.974) and 0.37 m s−1 (0.991), respectively, with correlation coefficients > 0.99.

Comments on “Wind Gust Detection and Impact Prediction for Wind Turbines”

“…While their paper mentions cross-correlation (CC) and wavelet-based optical flow (WOF) algorithms as alternative ways to obtain such wind fields, statements about these two methods are inaccurate and do not consider the latest published research on this subject. In this note, we correct statements in [1] based on our recent work [3,4] and put the CC, OF, and 2D-VAR methods into perspective. We contend that the WOF and 2D-VAR algorithms are in fact similar and that motion estimation methods should not be discounted for remote wind sensing applications such as the one they developed.…”

“…The quoted statements above ignore the abundance of published evidence of the strong skill of motion estimation algorithms and dismiss the approach for use in meteorological applications [3][4][5][7][8][9]. The literature shows that two numerical techniques (cross-correlation (CC) and wavelet-based optical flow (WOF)) are capable of extracting horizontal 2D-2C vector wind fields from near-horizontal aerosol lidar scans.…”

“…While this statement is correct in that the CC approach has limitations, as does any measurement technique, it fails to acknowledge the vastly more important point that the algorithm works well over a wide range of wind conditions. This includes moderate and strong winds, when non-uniformity in wind direction is small compared to light wind conditions that may occur in boundary layers driven exclusively by convection [4]. Zhou et al [1] focused on detecting and tracking gusts in the context of wind power production.…”

Comparison of an analog direct detection and a micropulse aerosol lidar at 1.5 - μ m wavelength for wind field observations—with first results over the ocean

“…14 The REAL has resulted in impressive aerosol imagery that can be used with motion estimation algorithms to deduce vector wind fields. [18][19][20] The REAL transmitter, however, is large and challenging to maintain. It places very significant limitations on the embodiment of the entire instrument, sets substantial requirements for operation, and therefore reduces deployment opportunities.…”

“…21 Both instruments performed 45-deg sector scans covering ½−15;30deg azimuth at 2-deg elevation in about 13 s for the REAL and 12 s for SAMPLE. Wind estimates were retrieved from both instruments using the two motion estimation algorithms that were previously validated on the REAL against an independent Doppler lidar: cross-correlation (software named Gale) 19 and wavelet-based optical flow (software named Typhoon). 20 At the time of the side-by-side experiment, SAMPLE was not capable of quickly returning the telescope pointing direction to the starting angle of each scan, a move employed by the REAL beam steering unit and referred to as "fly-back."…”