针对现有的地物回波识别算法在“新一代天气雷达建设业务软件系统开发及应用（ROSE）”前期评估中存在的地物回波漏判和气象回波误判问题，提出了地物回波识别改进算法。新算法在地物回波识别前对径向速度和速度谱宽资料作了处理，改善了因采用双PRF技术造成的径向速度资料质量下降对地物回波识别的影响，并增加了回波强度与径向速度处理阈值、回波强度垂直梯度参数，调整了回波强度沿径向的库间变化的判定阈值，减小了对气象回波的误判。利用SA、SB和CB雷达共432个地物与降水回波体扫资料进行数据统计和典型个例分析，对比算法改进前后的识别效果，结果表明：径向速度处理改善了径向速度资料质量，使地物回波的平均径向速度接近于零，将地物回波识别命中率提高了6.25%；阈值的设定和特征参数的改进，改善了对气象回波的误判，虚警率降低了1%~2%。
Ground clutter is an unpredictable event which often contaminates the radar products. In order to get high-quality radar base data and post-secondary radar exported products, ground clutter must be suppressed effectively. In this paper, according to the missed alarm rate of ground clutter and the false alarm rate of meteorological echo in the old ground clutter identification algorithm, which have been assessed in ROSE (Radar Operational Software Engineering) system, we propose an improved method. The radial velocity and spectral width are simply processed before the ground clutter is identified, so that the radial velocity quality in dual PRF mode is improved, and the missed alarm rate of ground clutter is reduced. In addition, echo intensity and radial velocity processing thresholds are added to the new algorithm; The threshold of numbers changes in slope within the gate-to-gate echo intensity is adjusted, and the vertical gradient parameter of reflectivity is increased. Four hundred and thirty-two ground clutter echo and meteorological echo data have been used to analyze and compare the identification effect of the new and old algorithm. The results show that the average radial velocity of ground clutter in dual PRF mode is close to zero after processing, and the accuracy of ground clutter identification is improved by 6. 84%. The added thresholds and the improvements of the characteristic parameters also could decrease the false alarm rate of meteorological echo about 1% to 2%. However, the new ground clutter identification algorithm fails to solve the false recognition problem completely. Meteorological echo, which have no radial velocity data (especially 230 km outside) and have great change in horizontal and vertical echo intensity, is very difficult to distinguish from ground clutter. This also needs to continue to do more work in the future. In particular, we could increase the new probe measurement, such as using dual polarization radar.
[1]Collier C G, Lovejoy S, Austin G L. 1980. Analysis of bright bands from 3-D radar data[C]//the 19th Conference on Radar Meteorology, Amer Meteor Soc, 44-47.
[2]Hogg W D. 1978. Quality control and analysis of an archive of digit al radar data[C]//the 18th Conference on Radar Meteorology, Amer Meteor Soc, 150-154.
[3]Holleman I, Beekhuis H. 2003. Analysis and correction of dual PRF velocity data[J]. J Atmos Ocean Technol, 20(4):443-453.
[4]Joe P, May P T. 2003. Correction of dual PRF velocity errors for operation Doppler weather radars[J]. J Atmos Ocean Technol, 20(4):429-442.
[5]Kessinger C, Ellis S, Andel J V. 2003. The radar echo classifier:a fuzzy logic algorithm for the WSR-88D[C]//31rd Conf Artificial Applications to Environmental Science, Amer Meteor Soc, Long Beach.
[6]Kessinger C, Ellis S, Andel J V, et al. 2005. The AP ground clutter mitigation scheme for the WSR-88D[C]//31st Conf on Radar Meteorology, Seattle Washington, Amer Meteor Soc, 526-529.
[7]Lakshmanan V, Hondl K, Stumpf G, et al. 2003. Quality control of weather radar data using texture features and a neural network[C]//5th International Conf on Adv in Patt Recogn.
[8]Mueller E A, Sims A L. 1975. Statistics of High Radar Reflectivity Gradients[C]//16th Radar Meteorology Conf, Amer Meteor Soc, 401-403.
[9]Rosenfeld D, Amitai E, Wolff D B. 1995. Classification of rain regimes by the three-dimensional properties of reflectivity fields[J]. J Appl Meteor, 34:198-211.
[10]Steiner M, Smith J A. 2002. Use of three-dimensional reflectivity structure for automated detection and removal of nonprecipitating echoes in radar data[J]. J Atmos Ocean Technol, 19:673-686.
[11]Ge Runsheng, Liu Enqing. 1989. An estimate of detectivity of CAMS Doppler weather radar[J]. Journal of Academy of Meteorological Science, 4(2):133-140.<br/>葛润生, 刘恩清. 1989.CAMS多普勒天气雷达探测能力的估算[J].气象科学研究院院刊, 4(2):133-140.
[12]Jiang Yuan, Liu Liping, Zhuang Wei. 2009. Statistical characteristics of clutter and improvements of ground clutter identification technique with Doppler weather radar[J]. J Appl Meteor Sci, 20(2):203-213.<br/>江源, 刘黎平, 庄薇. 2009.多普勒天气雷达地物回波特征及其识别方法改进[J].应用气象学报, 20(2):203-213.
[13]Liu Liping, Wu Linlin, Yang Yinming. 2007. Development of fuzzy-logical two-step ground clutter detection algorithm[J]. Acta Meteor Sinica, 65(2):252-260.<br/>刘黎平, 吴林林, 杨引明. 2007.基于模糊逻辑的分步式超折射地物回波识别方法的建立和效果分析[J].气象学报, 65(2):252-260.
[14]Li Feng, Liu Liping, Wang Hongyan, et al. 2014. Identification of ground clutter with C-band Doppler weather radar[J]. J Appl Meteor Sci, 25(2):158-167.<br/>李丰, 刘黎平, 王红艳, 等. 2014. C波段多普勒天气雷达地物回波识别方法[J].应用气象学报, 25(2):158-167.
[15]Shi Zhao, He Jianxin, Li Xuehua, et al. 2013. Analysis and application of dual polarization parameters about ground clutter from X-band weather radar[J]. Plateau Meteor, 32(5):1478-1484. DOI:10. 7522/j. issn. 1000-0534. 2012. 00137.<br/>史朝, 何建新, 李雪华, 等. 2013. X波段天气雷达地物回波的双偏振参量特征分析及应用[J].高原气象, 32(5):1478-1484.
[16]Wang Dan, Wang Gaili, Liu Liping, et al. 2014. Comparisons analysis on short-term precipitation between the radar-based extrapolation and the meso-scale numerical model weather prediction[J]. Plateau Meteor, 33(3):811-822. DOI:10. 7522/j. issn. 1000-0534. 2013. 00038.<br/>王丹, 王改利, 刘黎平, 等. 2014.基于雷达回波外推和中尺度模式预报的短时降水对比分析[J].高原气象, 33(3):811-822.
[17]Yang Chuan, Liu Liping, Hu Zhiqun, et al. 2012. An algorithm for chaos radial velocity identifying and processing in C-band Doppler radars running in the dual PRF mode[J]. Acta Meteor Sinica, 70(4):875-886.<br/>杨川, 刘黎平, 胡志群, 等. 2012. C波段多普勒雷达双PRF模式速度混淆区识别和处理方法研究[J].气象学报, 70(4):875-886.
[18]Yang Yin, Zhu Keyun, Zhang Jie, et al. 2015. Study on Doppler radar data assimilation over complex terrain[J]. Plateau Meteor, 34(5):1495-1501. DOI:10. 7522/j. issn. 1000-0534. 2014. 00059.<br/>杨银, 朱克云, 张杰, 等. 2015.复杂地形下多普勒雷达资料同化的研究[J].高原气象, 34(5):1495-1501.
[19]Zhuang Wei, Liu Liping, Yu Yanqun, et al. 2012. Improvement of the fuzzy logic technique for identifying ground clutter and its verification[J]. Acta Meteor Sinica, 70(3):576-584.<br/>庄薇, 刘黎平, 余燕群, 等. 2012.雷达地物回波模糊逻辑识别法的改进及效果检验[J].气象学报, 70(3):576-584.
