Study On Wind Field Retrieval For Spaceborne Doppler Cloud And Rain Radar

Global three-dimensional wind information is essential for operational weather forecasting on all scales and at all latitudes,which plays a unique role in understanding and predicting weather and climate.Doppler cloud and rain radar based on polar orbiting satellite makes it possible to obtain global three-dimensional wind field with its unique Earth observation perspective and fast speed around the earth.In view of the problems and challenges of acquiring atmospheric wind field under spaceborne platform,the wind field retrieval methods of spaceborne Doppler cloud and rain radar were systematically studied in this thesis from the basic conditions of wind field retrieval based on the full investigation of the development of Spaceborne Radar and Doppler radar wind field retrieval methods at home and abroad,this paper systematically studies.The main contents and conclusions are as follows:(1)Based on the analyzing the factors which impact the estimation accuracy of Doppler weather radar spectral parameters,combined with the research of spaceborne cloud-rain radar equation,system signal-to-noise ratio,characteristics of spaceborne meteorological targets and inherent constraints of radar itself,the key parameters of the system suitable for spaceborne cloud-rain radar for the measurement of radial velocity are obtained,such as central frequency,peak power of transmitter,pulse width,pulse repetition frequency,etc.Through analysis and comparison of the velocity estimation precision of each method,it is found that the velocity estimation accuracy is greatly affected by the signal-to-noise ratio(SNR)in the case of low SNR,the estimation error decreases exponentially with the increase of SNR,and the Pulse Pair Processing(PPP)method is better than the FFT method.When the error decreases to a certain threshold,the estimation error tends to be stable with the increase of the SNR.For a SNR better than 5 d B and coherent accumulation number greater than 32,the error of velocity estimation is less than 1 m/s.And the estimation accuracy of the two methods increases with the increase of the number of coherent accumulation points.Ka-band can provide a more precise Doppler velocity estimation than Ku-band under the same SNR.By using Polarization Diversity Pulse Pair(PDPP)method,the Doppler dilemma can be overcome but with performance degradation,and the accuracy is lower than that of PPP method.The estimation performance of PDPP can be improved by changing the time interval of polarization pulse_HV and the range of_HV is between10 us and 80 us.(2)The contribution of three-dimensional wind components to radar radial velocity in the scanning strategies of along track,cross track and conical scanning under spaceborne platform is systematically analyzed,and the comparison of wind retrieval capabilities of each scanning strategy is given:along track scanning can only obtain two-dimensional data,cross track and conical scanning can obtain three-dimensional data;conical scanning has the best swath while along track scanning has the smallest;the vertical component of wind field can be retrieved by a single radar in the nadir-looking along track scanning;the three-dimensional component of wind field can be retrieved theoretically by a single radar under the conical scanning strategy or by a multi-beam radar system under the along track and conical scanning;the wind field component cannot be retrieved in the zero azimuth cross track scanning except for a case when the beam pointing angle is zero and the vertical component of wind can be obtained.(3)Based on velocity azimuth display(VAD)method,a wind field retrieval method for spaceborne Doppler radar is proposed,and a space geometric observation model is established.The azimuth angle and first harmonic product term introduced to radial velocity is found during the analysis and deduction of VAD wind field retrieval under spaceborne conical scanning,and analyzed and the coefficient of the product term is derived.The performance of VAD wind retrieval under spaceborne application is validated based on simulated wind data and real data of High-Altitude Wind and Rain Airborne Profiler(HIWRAP)respectively,which shows the average horizontal wind field with higher vertical resolution is obtained.(4)The geometric model of dual-beam observation for spaceborne radar wind retrieval is designed.Constrained by the minimum spatial and temporal error of dual-beam observation,the key scanning parameters are optimized and determined,and a good coverage of the same area by the dual-beam is achieved at most azimuth angles.The dual-beam wind retrieval equation for spaceborne radar is established and solved.For a simulated typhoon field,the retrieval error is large when the SNR is low but the results still maintains the shape of the original typhoon wind field,and the root mean square errors(RMSE)of the retrieval,,are less than 2.03 m/s,2.01 m/s and 0.48 m/s respectively when the SNR is higher than 20d B.Due to the large difference of the beam pointing angle between the two beams of HIWRAP,the relative RMSE of dual-beam retrieval results for the Ku band and Ka band is 174%and135%respectively along the flight direction when the dual-beam wind field inversion method designed in this paper is applied.(5)The constraint equation of three-beam observation for spaceborne radar wind retrieval is deduced,and the optimized key scanning parameters are obtained.The validity of the designed scanning parameters is verified by analyzing the distance error of the three-beam observation footprint.The wind retrieval equation of three-beam is solved by the least square method.The method is validated by a simulated three-dimensional wind field of typhoon model.The results indicate that this method is of high precision and the RMSE of the retrieval,,are less than 0.486 m/s,0.742 m/s and 0.265 m/s,respectively.The retrieval error is less than5%with a small rainfall rate,and is smaller and smaller as the increase of rainfall rate,and the error is less than 1%when the rainfall rate reaches 18mm hr^-1.