Study On Wind Field Over Complex Terrain Obtained By Bistatic Doppler Wind Lidar And CFD Model

Understanding the details of micro-scale wind fields over complex terrain is important for military action,air quality evolvement,wind energy resource assessment,boundary layer meteorology,and so on.However,Due to the influence of topographic factors and thermal effects,the distribution characteristics of wind field over complex terrain are relatively complex.How to obtain accurate and reliable wind field over complex terrain is an urgent problem to be solved.The purpose of this paper is to obtain micro-scale wind field over complex terrain,and the main work is carried out from two aspects of observation and numerical simulation.Firstly,aiming at the limitation that monostatic doppler wind lidar can only detect the horizontally homogeneous wind,a feasibility study on wind field observation over complex terrain using bistatic doppler wind lidar is conducted.For this,a simulation system of bistatic doppler wind lidar was established.This system can directly detect the wind over complex terrain,eliminating the influence of artificial hypothesis.Due to the long scan period of bistatic wind lidar in the studied region,we mainly used it to detect wind profiles in strategic locations.Secondly,we combined the CFD(Computational Fluid Dynamics)model and outputs of mesoscale model,successfully simulated the micro-scale wind field over complex terrain.Finally,in order to improve the accuracy of simulated wind field,we applied the data assimilation(Newtonian Relaxation,Nudging)technique into the CFD model.Through integrating the observational data of bistatic doppler wind lidar and automatic meteorological station into the CFD model,the accuracy of wind field over complex terrain can be improved.The primary works and conclusions are listed below:(1)The detection capability of bistatic doppler wind lidar was studied,laying a foundation for building the system of bistatic doppler wind lidar.Firstly,using the improved volume scattering model,the equations of bistatic doppler wind lidar were established,and then the echo signal for incoherent(direct-detection)bistatic doppler wind lidar was provided.Secondly,according the results of signal to noise ratio(SNR),we chose 1064 nm as the wavelength of bistatic doppler wind lidar,200 as the number of pulse accumulation,2km as the baseline distance,and 3mrad as the field of view(FOV).Finally,according the relationship between doppler velocity error and SNR,the area where the SNR>26.5d B(that is,the doppler velocity error is less than 5%)was considered as the effective detection area of bistatic doppler wind lidar.For the wind lidar system with one active lidar and two passive lidar,it can effectively detect the cube area of 4km×4km×4km.(2)The simulation system of bistatic doppler wind lidar was established,and the application of wind retrieval was conducted.Firstly,the frequency discrimination system of bistatic doppler wind lidar was simulated.By detecting the frequency of backward-scattering and lateral-scattering echo signals,the doppler velocities of active and passive lidars were retrieved.Secondly,referring to the doppler relationship of bistatic wind lidar and the wind synthesizing method of multiple radars,we established the wind synthesizing method of bistatic doppler wind lidar.Thirdly,an equilateral triangle distribution of the active and passive lidars was selected as the layout of the bistatic doppler wind lidar system on the basis of the uncertainty of synthetic wind field.Fourthly,using high-resolution meteorological element profiles of Nanjing in 2015,we statistically analyzed the wind velocity error of bistatic doppler wind lidar.The results show that the wind velocity error at 4km below was less than 1m/s,and the relative error was less than 5%,meeting the requirements of atmospheric wind detection.Finally,using two typical wind field models over complex terrain,we compared the wind retrieval error between monostatic and bistatic doppler wind lidars.The results show that,due to the scanning mode,such as VAD(Velocity-Azimuth Display),DBS(Doppler Beam Scanning),etc.,is based on the assumption that the horizontal wind field is uniform,so the wind retrieval error of monostatic wind lidar was huge.As for the bistatic wind lidar,it can directly detect the wind over complex terrain,eliminating the influence of artificial hypothesis,so its wind retrieval error was small.(3)We combined the CFD model and outputs of mesoscale model,successfully simulated the micro-scale wind field over complex terrain,making up for the deficiency of bistatic doppler wind lidar.In order to achieve accurate mesoscale wind field,the sensitivity analysis of mesoscale model's physical schemes and grid scales were conducted.Hence,the best performance of mesoscale simulations is set as the boundary condition of CFD model in the Chong li Mountain region.To solve the problem of huge terrain differences between the mesoscale and CFD models,we improved the coupling method that based on geometric coordinate system,and couples these two models based on terrain following coordinate system.According to the simulation results,wind velocity distribution on the CFD boundary using coupling method that based on terrain following coordinate system is more realistic,and it has a better performance than that based on geometric coordinate system.(4)A data assimilation system,was applied in the CFD model,and the simulation accuracy of wind field over complex terrain was improved by integrating the data of bistatic doppler wind lidar.Firstly,we applied the data assimilation(Newtonian Relaxation,Nudging)technique into the CFD model by modifying the governing equation.Secondly,through integrating the wind data of bistatic doppler wind lidar,automatic meteorological station,etc.,into the CFD model,we conducted the experiments on the data assimilation.Results show that CFD wind data at assimilated site was nudged towards the observation value by the influence of data assimilation.Following the control equations of fluid mechanics and the whole conservative system(mass,momentum and energy conservations),the CFD wind data at other places were modified somewhat towards the observation values.There was a certain relationship between wind direction and the impact area of data assimilation.From the difference between the simulations of CFD model with and without assimilation,it was quite obvious that the wind field upstream and downstream of the assimilation site was strongly modified by Nudging.Finally,through integrating the observation data of automatic meteorological station into the CFD model,we conducted an experiment on the data assimilation,which further verified the feasibility of data assimilation.