Abstract:Pulsed Doppler wind lidars (PDWL) have been extensively used in order to study the atmospheric turbulence. Their ability to scan large areas in a short period of time is a substantial advantage over in-situ measurements. Furthermore, PDWL are capable to scan horizontally as well as vertically thus providing observations throughout the atmospheric boundary layer (ABL). By analysing PDWL observations it is possible to identify large turbulent structures in the ABL such as thermals, rolls and streaks. Even though several studies have been carried out to analyse such turbulent structures, these studies examine peculiar cases spanning over short periods of time.For this study we analysed the turbulent structures (thermals, rolls, streaks) over Paris during a two-months period (4 September &#8211; 6 October 2014, VEGILOT campaign) observed with a PDWL installed on a 70&#160;m tower in Paris city centre. The turbulent radial wind field was reconstructed from the radial wind field of the horizontal surface scans (1&#176; elevation angle) by using the velocity azimuth display method. The VEGILOT campaign provided 4577 horizontal surface scans, hence for the classification of the turbulent structures we developed an automatic method based on texture analysis and machine learning of the turbulent radial wind fields. Thirty characteristic cases of each turbulent structure types were selected at the learning step after an extensive examination of the meteorological parameters. Rolls cases were selected at the same time that cloud streets were visible on satellite images, streaks cases were selected during high wind shear development near the surface and thermals case were selected when solar radiation measurements in the area were high. In addition, sixty cases of &#8220;others&#8221;, representing any other type of turbulence, were added to the training ensemble. The analysis of errors estimated by the cross-validation shows that the K-nearest neighbours&#8217; algorithm was able to classify accurately 96.3% of these 150 cases. Subsequently the algorithm was applied to the whole dataset of 4577 scans. The results show 52% of the scans classified as containing turbulent structures with 33% being coherent turbulent structures (22% streaks, 11% rolls).Based on this classification, the physical parameters associated with the different types of turbulent structures were determined, e.g. structure size, ABL height, synoptic wind speed, vertical wind speed. Range height indicator and line of sight scans provided vertical observations that illustrate the presence of vertical motions during the observation of turbulent structures. The structure sizes were retrieved from the spectral analysis in the transverse direction relative to the synoptic wind, and are in agreement with the commonly observed sizes (a few 100&#160;m for streaks, a few km for rolls).

Processing of turbulence-layer wind speed with Generalized SCIDAR through wavelet analysis

Generalized SCIDAR Measurements at San Pedro Mártir. II. Wind Profile Statistics

A New Wind Speed Retrieval Method for an Ocean Surface Using the Waveform Width of a Laser Altimeter

Estimation of the Motion-Induced Horizontal-Wind-Speed Standard Deviation in an Offshore Doppler Lidar

Wind Directions Retrieval from SAR Image Combining the Spatial and Spectral Domains: A Novel Approach Using a Modified Circle Median Filter

Arrival-Time Detection in Wind-Speed Measurement: Wavelet Transform and Bayesian Information Criteria

Wavelet analysis of atmospheric turbulent data

Turbulence Detection in the Atmospheric Boundary Layer using Coherent Doppler Wind Lidar and Microwave Radiometer

Accuracy of estimation of the turbulence energy dissipation rate in the atmospheric boundary layer from measurements of the radial wind velocity by micropulse coherent Doppler lidars

Wave and turbulence separation using dynamic mode decomposition

Machine-learning identification of the variability of mean velocity and turbulence intensity for wakes generated by onshore wind turbines: Cluster analysis of wind LiDAR measurements

Adaptive Optics Parameters connection to wind speed at the Teide Observatory

Directly Measuring Atmospheric Turbulence Parameters Using Coherent Doppler Wind Lidar

Statistical study of coherent turbulent structures properties observed by a Doppler lidar over Paris during two months

Evaluating wind profiles in a numerical weather prediction model with Doppler lidar

A novel integrated algorithm for wind vector retrieval from conically scanning scatterometers

A versatile scanning method for volumetric measurements of velocity and density fields

Detection of clouds in multiple wind velocity fields using ground-based infrared sky images

A Spatiotemporal Estimation Framework for Real-World LIDAR Wind Speed Measurements

Profiling the molecular destruction rates of temperature and humidity as well as the turbulent kinetic energy dissipation in the convective boundary layer

Turbulence statistics estimation across a step change in roughness via interpretable network-based modelling