Doppler Asymmetric Spatial Heterodyne Technique For Wind Detection In The Upper Atmosphere

The wind detection in the upper atmosphere plays an important role in understanding the dynamics and actinochemistry,building the atmospheric dynamic model,offering the long term weather forecast and guaranteeing the development of aeronautics and astronautics.Regarding the atmospheric airglow as light source,the Doppler effect is used to retrieve wind parameters in passive wind detection.Typical passive wind detection technique includes Fabry-Perot interference technique and Michelson interference technique.Recently,the Doppler Asymmetric Spatial Heterodyne(DASH)spectroscopy becomes the research focus depending on its wide field of view,large etendue,high spectrum resolution,static measurement and multiline capability.Combining spatial heterodyne spectroscopy and Michelson interference technique,wind speed is retrieved through change of the interference phase in large Optical Path Difference(OPD).The foundation of DASH technique has been built by designing the redline DASH demonstration instrument called REDDI.To achieve high stability,the interferometer uses quasi-common path configuration through a Koster's double-reflection prism.However,the penalty for quasi-common path configuration is small OPD interval and low phase sensitivity.On the other hand,the retrieving of wind speed is considered immune to the phase error by differencing the absolute phase.Therefore at present the data processing is conducted without considering the error.The wind precision of the ground-based instrument is influenced by small phase sensitivity and the data processing error both.Based on the factors above,the double-arms LODI(Large Offset DASH Interferometer)instrument is developed in this paper.By designing the wind simulator and building the system setup,the wind simulating detection is achieved.During the data processing,the optimization selection of the window parameters and the correction method of the absolute phase drift are proposed.The main research contents of this thesis are as follows:1.The wind detection principle based on DASH technique is analyzed.Starting on the choice of detection target,character of the target and demand of the interferometer are discussed.Based on the theoretic model of DASH interferometer,key parameters aiming to wind detection are derived and analyzed in order to clear out the basis of system designing.2.The LODI parameter is proposed according to the DASH technique.The interferometric system is simulated and designed.The drift of Littrow wavenumber,absolute phase and wind speed error caused by the error of optical element(beam splitter,field-widened prism and gratings),system noise and temperature change of the environment is simulated and analyzed quantitatively.After that the detailed parameters of the interferometer system is designed according to the requirement.3.The wind retrieval algorithm of DASH interferometer is studied.By comparing two different methods used to calculate absolute phase,the conclusion that the latter one is more suitable to achieve phase and retrieve wind speed is got.The calculation error during the data processing is analyzed under the presupposition.As an external variable,window function leads to the data processing error.The optimization selection method of the window parameters is proposed.It is indicated that by choosing the Nuttall window with the FWHM of 2,the wind speed error caused by the window parameters can be minimized.4.The experiment of wind simulation detection based on LODI is discussed.According to the principle of wind simulation,the detailed machine design scheme and the error analysis are given.It is shown that the total wind simulating error is approximately 1.3%,which is due to the angle and measurement error.Centered on the DASH interferometer,the experimental setup is built to detect the wind speed.During the optimization selection analysis of the window parameters,it is shown that the rectangle window leads to a wind speed of 1.75 m/s in the OPD interval,while the error is only 0.75 m/s under Nuttall window.With the increase of window width,the wind speed error increases.Choosing the narrow Nuttall window at FWHM=2,the wind speed error decreased 1.5 m/s.From 20 groups of single measurement at a given simulated wind speed,it is indicated that the retrieved speed error is 6.1 m/s,which is due to the phase drifts.The absolute phase error is corrected by linearly fitting the absolute phase series.The wind speed error is improved with 28 m/s.According to the data correction above,24 groups of wind simulation detection from 34.19 m/s to 78.63 m/s are achieved.The corresponding wind retrieving error and uncertainty are calculated and finally a wind precision of 2.92 m/s is obtained.