Accuracy of Doppler LIDAR Estimates of Vertical Wind Shear and Turbulence From a Moving Platform

One of the most renowned problems currently investigated in science is the uncertainty associated with climate model projections. It has been addressed in reports written by members of IPCC (Intergovernmental Panel on Climate Change), as well as in other academic literature, that a large uncertainty still remains in climate model outcomes despite improvements in model resolution and the model physics. These issues—heavily outlined in IPCC reports—are largely due to a lack of observations, both spatially and temporally, and a lack of understanding in the physical processes that govern certain regions of interest. As a result, the scientific focus lately has shifted towards improving our understanding of the physical processes that lead to climate change, and, expanding the observational network to improve upon our understanding and reduce model uncertainty.

Despite a clear improvement in both the quantity and quality of observations, it still remains difficult to investigate how offshore processes, such as air-sea interactions and freshwater loading, impact the changes we see in climate. It has been noted on several occasions that the difficulty in acquiring offshore measurements is largely due to offshore expenses, the hardships of offshore fieldwork, and the vast coverage of the world's oceans as compared to onshore. This is viewed as being particularly problematic when considering that the world's oceans largely influences global circulation. Recent developments however, have been put forth to increase the amount of permanent offshore platforms with sophisticated instrumentation, for instance, through the deployment of Wind Sentinels—a large self-powered platform—fitted with meteorological sensors and wind LIDARs (LIght Detection And Ranging). Examples of such deployments are off the coast of Virginia Beach, VA, Lake Michigan, and until recently, Portland, OR.

The deployment of a Doppler LIDAR offshore, though not frequent, has occurred numerous times prior to the deployment of Wind Sentinel platforms. It has been known, particularly for marine-borne vessels, that LIDAR measurements incorporate additional Doppler shifts into wind data due to platform-wave interactions. Many researchers have explained that a wind measurement, when properly removed of motion, accurately describes the wind as though the platform to which its' secured were stationary. However, evidence has shown through case studies conducted by members of the University of Agder in 2011, and through results of the XPIA field campaign at BAO (Boulder Atmospheric Observatory) in 2015, that despite motion compensation, errors still reside in wind LIDAR data. This poses a greater concern when considering both the current lack of understanding of offshore meteorology, and the uncertainty in Wind LIDAR data present after motion compensation.

The work presented examines additional causes of uncertainty present after motion compensation, as well as features in the wind data during moments of high seas. The first part of the analysis examines a simulation of a simple wind field that obeys the surface layer equation (i.e. 10% of boundary layer) under neutral conditions. It is shown that when modeling the wind profile under different platform orientations, that a strong relationship is formed between the change in the wind measurement due to the platform tilt, the angle that the platform is tilted, the height of the wind measurement, and the prescribed shear of the wind profile. Results show that if the wind profile satisfies the condition of being in the surface layer, and under neutral conditions, that only the angular tilt of the platform is needed to determine the difference of the LIDAR measurement. The second part of the analysis introduces in a broader context, a wind profile with both horizontal variability and a complex vertical structure. In much the same way, the simulated wind field is modeled as LIDAR data; however, the error due to horizontal variability had to be taken into account to derive wind speed, and, a separate relationship was derived to determine the difference in the LIDAR wind measurement due to platform orientation. Results from this numerical experiment show that a near linear relationship can be derived between the difference in the LIDAR measurement due to platform orientation and the platform tilt, with the slope related to vertical wind shear. As a final analysis, the affect of high seas was examined on turbulent statistics of wind data. It is shown that when the tangential velocities exceeded the natural variability of the wind, a high bias was present in the turbulent dissipation rate. Once the wind was motion compensated, the LIDAR data exhibited natural variations, as compared to nearby stationary LIDARs, with the vertical structure statistically changing the same way. However, it was also found that large gradients in turbulence skewed the result, thus showing the gradient in the turbulent dissipation rate to be different between the uncompensated and compensated profiles.