双频雷达通过估计降水的雨滴谱分布（Rain Drop Size Distribution, DSD）参数来估测降水， 相比多普勒天气雷达利用Z-I关系估测降水的方法， 估测结果更加准确。真实的雨滴谱分布随时空不断变化， 难以准确描述。本研究的目的是评估将其描述成Gamma分布模型对双频雷达估测降水造成的误差。研究中利用架设在广东龙门和西藏那曲的雨滴谱仪观测资料， 计算Ku/Ka双频雷达的反射率因子， 利用双频雷达算法反演降水率R和衰减系数k， 并将其与雨滴谱仪直接计算的结果进行比较， 探讨双频雷达反演算法中DSD模型对反演结果造成的影响， 并对比反演误差的地区差异性。研究结果表明: DSD参数对双频雷达反演降水的影响在那曲和龙门地区有一定的差异。从双频技术反演的质量加权的粒子直径（Dm）、 Ka波长衰减系数（kKa）、 Ku波长衰减系数（kKu）和降水率(R)来看， 当Gamma分布的形状因子μ分别取2、 3和4时， 龙门地区对四个参数的反演都能得到较为稳定准确的结果， 相对误差基本<10%。而就kKa和5~40 mm·h-1的降水率而言， 那曲和龙门有很大的不同， 在龙门当μ取3时相对误差最小， 都在0值附近波动， 而那曲则要需要μ值在4和6附近才能使得相对误差达到最小。
Compared with Doppler radar， which uses radar reflectivity-rainfall rate (Z-R) relationship to estimate precipitation rates， the dual-frequency radar can provide more accurate estimates of precipitation rates based on retrieving the Rain Drop Size Distribution (DSD) parameters.However， actual DSD spectra varies with time and space and is difficult to be described accurately.The purpose of this study is to evaluate the retrieval errors caused by DSD models adopted in Ku/Ka-band dual-frequency radar retrieval.Based on continuous summer DSD data from PARSIVEL disdrometer at Longmen and Naqu observatories， the radar effective reflectivity factors at the Ka and Ku bands are calculated.Then， the median mass-weighted diameter Dm， attenuation coefficients k and rain rates R obtained from the DSD estimated by dual-frequency techniques are compared with those directly computed from DSD spectra measured by the disdrometer to investigate the impact of DSD model on the dual-frequency retrieval techniques and compare the differences between different regions.Overall， a gamma DSD model with fixed μ values of 2， 3 or 4 could yield consistent and fairly accurate retrievals， although the impacts of DSD parameters on the dual-frequency radar retrieval technique in Naqu and Longmen regions are different.In Longmen region， the relative biases of the retrieved Dm， the Ka-band attenuation coefficient kKa， the Ku-band attenuation coefficient kKu and R are less than ±10% bias when the shape factor μ of the gamma distribution ranges from 2 to 4.In terms of kKa and rain rate of 5~40 mm·h-1， the Gamma DSD model with μ=3 in Longmen region could yield the best accuracy and the relative bias fluctuates around 0， while the Gamma DSD model with μ=4， 6 in the Naqu region could produce the smallest errors.
[1]Adhikari N B， Iguchi T， Seto S， al et， 2007.Rain retrieval performance of a dual-frequency precipitation radar technique with differential-attenuation constraint[J].IEEE Transactions on Geoscience & Remote Sensing， 45(8): 2612-2618.DOI: 10.1109/tgrs.2007.893555.
[2]Atlas D， Srivastava R C， Sekhon R S， 1973.Doppler radar characteristics of precipitation at vertical incidence[J].Reviews of Geophysics， 11(1): 1-35.DOI: 10.1029/RG011i001p00001.
[3]Barber P， Yeh C， 1975.Scattering of electromagnetic waves by arbitrarily shaped dielectric bodies[J].Applied Optics， 14(12): 2864-2872.DOI: 10.1364/AO.14.002864.
[4]Carbone R E， Nelson L D， 1977.The evolution of raindrop spectra in warm-based convective storms as observed and numerically modeled[J].Journal of the Atmospheric Sciences， 35(12): 2302-2314.DOI: 10.1175/1520-0469(1978)035<2302: TEORSI>2.0.CO; 2.
[5]Joss J， Gori E G， 1978.Shapes of raindrop size distributions[J].Journal of Applied Meteorology， 17(17): 1054-1061.DOI: 10. 1175/1520-0450(1978)0172.0.CO; 2.
[6]Le M， Chandrasekar V， 2014.An algorithm for drop-size distribution retrieval From GPM dual-frequency precipitation radar[J].IEEE Transactions on Geoscience & Remote Sensing， 52(11): 7170-7185.DOI: 10.1109/tgrs.2014.2308475.
[7]Liao L， Meneghini R， 2005.A study of air/space-borne dual-wavelength radar for estimation of rain profiles[J].Advances in Atmospheric Sciences， 22(6): 841-851.DOI: 10.1007/BF02918684.
[8]Liao L， Meneghini R， Tokay A， 2014.Uncertainties of GPM DPR rain estimates caused by DSD parameterizations[J].Journal of Applied Meteorology & Climatology， 53(11): 2524-2537.DOI: 10.1175/JAMC-D-14-0003.1.
[9]Mardiana R， Iguchi T， Takahashi N， 2004.A dual-frequency rain profiling method without the use of a surface reference technique[J].Geoscience & Remote Sensing IEEE Transactions on， 42(10): 2214-2225.DOI: 10.1109/TGRS.2004.834647.
[10]Meneghini R， Kozu T， Kumagai H， al et， 1992.A study of rain estimation methods from space using dual-wavelength radar measurements at near-nadir incidence over ocean[J].Journal of Atmospheric & Oceanic Technology， 9(4): 364-382.DOI: 10.1175/1520-0426(1992)0092.0.CO; 2.
[11]Marshall J S， Palmer W M K， 1948.The distribution of raindrops with size[J].Journal of Meteorology， 5(4): 165-166.DOI: 10.1175/1520-0469(1948)005<0165: TDORWS>2.0.CO; 2.
[12]Porcù F， D'Adderio L P， Prodi F， al et， 2014.Rain drop size distribution over the Tibetan Plateau[J].Atmospheric Research， 150(150): 21-30.DOI: 10.1016/j.atmosres.2014.07.005.
[13]Srivastava R C， 2010.Size distribution of raindrops generated by their breakup and coalescence[J].Journal of Atmospheric Sciences， 28(3): 410-415.DOI: 10.1175/1520-0469(1971)028<0410: SDORGB>2.0.CO; 2.
[14]Steiner M， Yuter S E， 1995.Climatological characterization of three dimensional storm structure from operational radar and rain gauge data[J].Journal of Applied Meteorology， 34(9)， 1978-2007.DOI: 10.1175/1520-0450(1995)034<1978: CCOTDS>2.0.CO; 2.
[15]Ulbrich C W, 1983.Natural variations in the analytical form of the raindrop size distribution[J].Journal of Climate and Applied Meteorology， 22(10): 1764-1775.DOI: 10.1175/1520-0450(1983)022<1764: NVITAF>2.0.CO; 2.
[16]Willis P T， 1984.Functional fits to some observed drop size distributions and parameterization of rain[J].Journal of the Atmospheric Sciences， 41(9): 1648-1661.DOI: 10.1175/1520-0469(1984)041<1648: FFTSOD>2.0.CO; 2.
[17]Wu Y， Liu L， 2017.Statistical characteristics of raindrop size distribution in the Tibetan Plateau and Southern China[J].Advances in Atmospheric Sciences， 34(6): 727-736.DOI: 10.1007/s00376-016-5235-7.
[18]常祎， 郭学良， 2016.青藏高原那曲地区夏季对流云结构及雨滴谱分布日变化特征[J].科学通报， 61(15): 1706-1720.DOI: 10.1360/N972015-01292.
[19]陈月， 李跃清， 范广洲， 等， 2019.青藏高原大气蕴含潜热时空分布特征研究[J].高原气象， 38(3): 460-473.DOI: 10.7522/j.issn.1000-0534.2018.00130.
[20]常姝婷， 刘玉芝， 华珊， 等， 2019.全球变暖背景下青藏高原夏季大气中水汽含量的变化特征[J].高原气象， 38(2): 227-236.DOI: 10.7522/j.issn.1000-0534.2018.00080.
[21]陈宝君， 李子华， 刘吉成， 等， 1998.三类降水云雨滴谱分布模式[J].气象学报， 56(4): 506-512.DOI: 10.11676/qxxb1998.046.
[22]牛生杰， 安夏兰， 桑建人， 2002.不同天气系统宁夏夏季降雨谱分布参量特征的观测研究[J].高原气象， 21(1): 37-44.DOI: 10.3321/j.issn: 1000-0534.2002.01.007.
[23]濮江平， 张伟， 姜爱军， 等， 2010.利用激光降水粒子谱仪研究雨滴谱分布特性[J].气象科学， 30(5): 701-707.DOI: 10.3969/j.issn.1009-0827.2010.05.018.
[24]濮江平， 张昊， 周晓， 等， 2012.对流性降水雨滴谱特征及其与雷达反射率因子的对比分析[J].气象科学， 32(3): 253-259.DOI: 10.3969/2012jms.0013.
[25]唐国强， 万玮， 曾子悦， 等， 2015.全球降水测量(GPM)计划及其最新进展综述[J].遥感技术与应用， 30(4): 607-615.DOI: 10. 11873/j.issn.1004-0323.2015.4.0607.
[26]王玉琦， 鲍艳， 南素兰， 2019.青藏高原未来气候变化的热动力成因分析[J].高原气象， 38(1): 29-41.DOI: 10.7522/j.issn. 1000-0534.2018.00066.
[27]王鹏飞， 李子华， 1989.微观云物理学[M].北京: 气象出版社， 406-408．
[28]吴亚昊， 刘黎平， 周筠琚， 等， 2016.雨滴谱的变化对降水估测的影响研究[J].高原气象， 35(1): 220-230.DOI: 10.7522/j.issn.1000-0534.2014.00093.
[29]许丽生， 陈洪滨， 丁继烈， 等， 2014.非球形粒子光散射计算研究的进展综述[J].地球科学进展， 29(8): 903-912.DOI: 10. 11867/j.issn.1001-8166.2014.08.0903.
[30]张培昌， 杜秉玉， 戴铁丕， 2000.雷达气象学[M].北京: 气象出版社， 172-175.
[31]张鸿发， 徐宝祥， 蔡启铭， 1989.由雨滴谱型订正雷达测量降水的一种方法[J].高原气象， 8(1): 75-79.
