Research On Data Quality Control Methods Of High-resolution Weather Radar Network And Rainfall Retrieval In Complex Terrain

Continuous rainfall is one of the main factors causing the geological disasters such as landslides,mudslides and collapses.The data of precipitation with high resolution observed in complex mountainous and frequent geological disasters areas can provide a strong support for accurately rainfall estimation and rainfall-induced geological disasters forecasting.At present,the most effective way for tracking and monitoring of small and medium-scale precipitation systems is to use weather radar(rainfall radar).The networked radar composed of several mobile X-band all-solid-state weather radars can offer better solutions to the problems that afflict the new-generation weather radar network of China,such as the near-ground blind zone,the lack of data and the low resolution observation caused by earth curvature,beam block and the beam broadening in the key observation area.A series of data quality control algorithms based on networked radar are proposed in this paper.The proposed algorithms are well to take advantage of multi-view observation and data complementation of the observation model by networked radar.It provides solutions for typical error sources of radar data such as ground clutter,signal attenuation and beam block.It also helps to improve the accuracy and reliability of networked radar rainfall estimation,and increase the deep and wide application of the X-band network radar data in the meteorological,geological and hydrological research.The main contents and conclusions are as follows:(1)The potential factors that may affect the detection performance of the X-band solid-state weather radar(XY-A radar)are analyzed and discussed.In the detection performance and data reliability analyses,a collocated X-band klystron weather radar(WSR-98 XD radar)is used as the reference for evaluating the measurement accuracy of the XY-A radar.The results show that although the detection performance of the XY-A radar is limited by the transmit power,it has a high accuracy and reliability of medium-to-high intensity weather radar data(the reflectivity factors >20d Bz)when compared with the near-by WSR-98 XD radar.(2)According to the difference scattering characteristics of precipitation and ground clutter,a ground clutter recognition algorithm based on fuzzy logic and networked radar data is proposed.Combined the recognition results with the frequency domain clutter filtering technique,a ground clutter suppression method is designed in this paper.The proposed clutter suppression method can obtain the clutter recognition result in real time and reserved the weather information.In addition,the proposed method can also reduce the misjudgment rate of superrefraction clutter.(3)Take the advantage of multi-view radar network,an algorithm(Network algorithm)is studied to correct the attenuation of reflectivity factor.The simulation test of the algorithm shows that the signal fluctuation,systematic error and random error will not affect the results of corrected reflectivity factor.The improvement of the radar network data resolution is beneficial to make the corrected result closer to the simulated true-value.The Network algorithm and the traditional ZPHI algorithm are applied in radar data from out-field experiment on June 7,2017.The quantitative comparison with S-band radar observation data show that the Network algorithm is beneficial to correct the precipitation attenuation and improve the data quality of single polarization radar network.(4)A new algorithm based on the Network algorithm is proposed to solve the problem of radar partial beam block in complex terrain.The research results show that the proposed algorithm can effectively compensate the strip-shaped weak echo caused by the partial beam block,and increase the continuity of radar echo.In addition,the mosaic products of reflectivity factor from the networked radar is benefit to solve the data missing problem in null cone and complete beam block area.It is beneficial to improve the quality and continuity of radar data in the common area of radar network.(5)Firstly,the optimal R-Zh relationship and quantitative precipitation estimation(QPE)method based on the radar-rain-gauge sets is described.Secondly,the QPE estimated by the original and processed networked radar data are evaluated and analyzed by using the gauges network measurement and the merged satellite-gauge precipitation products from the China Meteorological Information Center(NMIC),respectively.The results show that networked radar data quality control algorithms can improve the accuracy of precipitation estimation and increase the continuity of the QPE field when the networked radars are observing in the complex terrain areas.