Corrections for Wind-Speed Errors from Sodar and Lidar in Complex Terrain

The quality of lidar and sodar wind estimates is generally judged through comparisons with mast-mounted instruments, and the resulting regressions. Evaluation of the relative merits of lidars versus sodars is complicated by the fact that lidars are generally placed close to a mast whereas sodars are generally placed some distance from a mast so that acoustic reflections off the mast are reduced. This leads to the two technologies, lidar and sodar, not being compared in similar situations. Differences arising from the two geometries can be expected to be larger in complex terrain, where the wind regime can vary significantly spatially. The current work explores these differences in moderately complex terrain. Lidar–mast comparisons are performed with the lidar close to an 80 m mast, and sodar–mast comparisons performed with the sodar 300 m from the mast. Systematic variations in estimated wind speed are found to occur with height, consistent with predictions from a simple flow model. When the lidar was moved to the sodar location, further from the mast, there were significant changes in the estimated wind speeds and a reduction in correlation with the mast-based wind speeds, as expected. However, the correlation between collocated lidar and sodar winds was high. This finding emphasizes that any comparison of two remote sensing instruments needs to be through similar experiments, and that differences in accuracy often reported for the lidar and sodar technologies are likely to be contaminated due to poor comparison configurations. A method was devised to simulate the sodar being collocated with the mast, by using the lidar–sodar measurements and the lidar–mast measurements. It was found that there was then no statistically detectable difference between lidar–mast regressions and sodar–mast regressions for the particular lidar and sodar tested. Both remote sensing instruments were also found to be good estimators of Weibull parameters, as compared with those derived from mast data. The conclusion is that the sodar measured the winds above the sodar with a similar accuracy to the lidar measuring winds above the lidar.