EXPERIMENTAL STUDY ON INFLUENCE OF DIFFERENT ASSIMILATION SCHEMES ON ENSEMBLE FORECAST OF TORRENTIAL RAIN IN SOUTH CHINA
-
摘要: 基于全球集合预报系统(GEFS)资料，利用WRF中尺度模式及GEFS动力降尺度获取区域集合预报初值场，通过对同化后的分析场进行模式积分实现华南前汛期区域集合预报。对2019年6月10日的一次华南前汛期暴雨过程进行不同同化方案的试验：混合同化(Hybrid)、三维变分(3Dvar)、集合卡尔曼滤波(EnKF)和对比试验(Ctrl)四组试验的对比分析，探讨具有不同背景误差协方差矩阵的同化方案对区域集合预报集合扰动和集合离散随时间演变特征的影响，评估不同试验的降水模拟效果。(1) Hybrid对模式初始场有较好的改善作用，而3DVar和EnKF对初始场的改善作用不明显。(2) 对风场、温度场和湿度场，在前期预报中Hybrid的预报误差小于3DVar和EnKF，在中后期的预报中，3DVar和EnKF的预报误差得到改善，且好于Hybrid。同样，集合扰动能量，Hybrid和Ctrl在前期预报发展好于3DVar和EnKF，而在中后期的预报3DVar和EnKF好于Hybrid和Ctrl。(3) 从24 h累积降水评分中，整体上同化试验好于Ctrl，3DVar和EnKF好于Hybrid，且3DVar对大中雨级别的降水评分较好，而EnKF对暴雨以上级别的降水评分较好。(4) 对于集合统计检验分析，同化试验的AUC值都大于Ctrl的AUC值，24 h累积降水量阈值在10~100 mm的AUC值，3DVar最好；而125 mm阈值的AUC值，EnKF最好。Abstract: Based on the Global Ensemble Forecast System data, the present study uses the WRF model and the GEFS dynamic downscaling method to obtain the regional ensemble forecast initial states. Moreover, the assimilated analysis field is integrated to achieve the reginal ensemble forecast for a precipitation event during annually first rainy season in South China. Four tests, namely Hybrid, 3DVar, EnKF, and Ctrl, are carried out for a torrential rain process on June 10, 2019 in South China. We also explore the evolution characteristics of the ensemble disturbance and ensemble spread for the assimilation schemes with difference background error covariance matrices and evaluate the precipitation simulation performance of different tests. The results show that: (1) Hybrid can improve the initial field of the model, while 3DVar and EnKF failed. (2) For wind, temperature, and relative humidity, the forecast error of Hybrid in the early forecast is less than that of 3DVar and EnKF. In the middle and late forecast, the forecast error of 3DVar and EnKF is reduced and is smaller than that of Hybrid. As for the ensemble disturbance energy, Hybrid and Ctrl are better than 3DVar and EnKF in the early forecast, and 3DVar and EnKF are better than Hybrid and Ctrl in the middle and late forecasts. (3) According to the 24-hour cumulative precipitation scores, the assimilation test is better than Ctrl, 3DVar and EnKF are better than Hybrid; 3DVar scores the best in heavy and moderate rainfall and EnKF scores the best in torrential rain and above. (4) For 24-hour cumulative precipitation ensemble statistical analysis, the area under the curve value of the assimilation test is greater than that of Ctrl; 3DVar performs the best in the 10mm~100mm cumulative precipitation threshold and EnKF performs the best in the 125mm cumulative precipitation threshold.
-
图 4 2019年6月10日06时700 hPa、500 hPa和300 hPa高度层的三组同化试验集合平均湿度场增量(填色; 单位：%) 和温度场增量(等值线；单位：K)
图 8 2019年6月10日06时起报的三组同化试验700 hPa风、温度和湿度预报场的均方根误差与对比试验的差值随时间变化
a. U风；b. V风；c. 温度；d. 相对湿度。
表 1 本文使用的全球同化系统观测资料表
数据 数据描述 格式 1bamua.tm00.bufr_d AMSU-A NCEP-proc. br. temps BUFR 1bhrs4.tm00.bufr_d HIRS-4 1b radiances BUFR 1bmhs.tm00.bufr_d MHS NCEP-processed br. temp BUFR ssmisu.tm00.bufr_d DMSP SSM/IS 1C radiance data (Unified Pre-Proc.) BUFR gpsro.tm00.bufr_d GPS radio occultation data BUFR 
下载: 导出CSV
表 2 四组集合预报成员扰动初值的形成及特点
试验 初始扰动来源 初始扰动更新 背景误差协方差 更新后的扰动能量范围 Ctrl 全球集合预报场GEFS
动力降尺度未同化 未同化 0.299~11.015 3DVar 3DVar同化 静态背景误差协方差 0.305~10.185 EnKF EnKF同化 集合样本背景误差协方差 0.336~9.717 Hybrid Hybrid同化 静态背景误差协方差(25%)+集合
样本背景误差协方差(75%)0.315~11.069
下载: 导出CSV
表 4 2019年6月10日06时起报的四组试验的24 h累计降水量不同阈值的AUC值
降水量/mm Hybrid 3DVar EnKF Ctrl 10 0.697 0.738 0.735 0.690 25 0.679 0.740 0.728 0.677 50 0.669 0.745 0.731 0.656 100 0.701 0.899 0.898 0.701 125 0.763 0.776 0.911 0.770
下载: 导出CSV
表 5 第24时预报场的U风、V风、温度和相对湿度的概率均方根误差表
物理量 Hybrid 3DVar EnKF Ctrl T 0.047 88 0.068 76 0.115 99 0.053 31 V 0.075 19 0.082 78 0.081 34 0.075 22 U 0.049 47 0.052 41 0.072 44 0.048 88 RH 0.076 62 0.082 71 0.080 93 0.075 46
下载: 导出CSV
-
[1] 马旭林, 时洋, 和杰, 等. 基于卡尔曼滤波递减平均算法的集合预报综合偏差订正[J]. 气象学报, 2015, 73(5): 952-964. [2] TALAGRAND O. Assimilation of observations, an introduction[J]. J Meteor Soc Japan, 1997, 75(1): 191-209. [3] 曹小群, 黄思训, 张卫民, 杜华栋. 区域三维变分同化中背景误差协方差的模拟[J]. 气象科学, 2008(1): 8-14. [4] BUEHNER M. Ensemble-derived stationary and flow-dependent background-error covariances: Evaluation in a quasi-operational NWP setting[J]. Quart J R Meteor Soc, 2005, 131(607): 1 013-1 043. [5] EUGENIAKALNAY. 大气模式、资料同化和可预报性[M]. 北京: 气象出版社, 2005. [6] HAMILL T M, SNYDER C. A hybrid ensemble Kalman filter 3D-Variational analysis scheme[J]. Mon Wea Rev, 2000, 128(8): 2 905-2 919. [7] 张诚忠, 薛纪善, 黄燕燕, 等. 资料同化对2017年登陆广东沿海台风的短期降水与路径预报影响[J]. 热带气象学报, 2019, 35(5): 577- 586. [8] 王洪, 王东海, 万齐林. 多普勒雷达资料同化在"7.21"北京特大暴雨个例中的应用[J]. 气象学报, 2015, 73(4): 679-696. [9] 张明阳, 张立凤, 张斌, 等. 集合变分混合同化背景误差协方差流依赖性分析[J]. 气象科学, 2015, 35(6): 728-736. [10] 夏宇, 陈静, 刘艳, 等. GRAPES混合同化方法在青藏高原区域的初步试验[J]. 大气科学学报, 2018, 41(2): 239-247. [11] WANG X G, BARKER D M, SNYDER C, et al. A Hybrid ETKF-3DVAR data assimilation scheme for the WRF model Part Ⅰ: Observing system simulation experiment[J]. Mon Wea Rev, 2008, 136(12): 5 116-5 131. [12] WANG X G, BARKER D M, SNYDER C, et al. A Hybrid ETKF-3DVAR data assimilation scheme for the WRF model Part Ⅱ: Real observation experiments[J]. Mon Wea Rev, 2008, 136(12): 5 132-5 147. [13] XIA Y, CHEN J, ZHI X F, et al. Topographic dependent horizontal localization scale scheme in GRAPES-MESO hybrid En-3DVAR assimilation system[J]. J Trop Meteor, 2018, 24(2): 245-256. [14] 陈静, 薛纪善, 颜宏. 华南中尺度暴雨数值预报的不确定性与集合预报试验[J]. 气象学报, 2003, 71(4): 432-446. [15] TOTH Z, KALNAY E. Ensemble forecasting at NCEP and the breeding method[J]. Mon Wea Rev, 1997, 125(12) : 3 297-3 319. [16] 于永锋, 张立凤. 基于增长模繁殖法的集合预报初始扰动饱和分析[J]. 大气科学, 2005, 29(6): 113-122. [17] BUIZZA R, PALMER T N. The singular-vector structure of the atmospheric global circulation[J]. J Atmos Sci, 1995(52) : 1 434 -1 456. [18] HOFFMAN R N, KALNAY E. Lag ged average forecasting an alternative to Monte Carlo forecasting[J]. Tellus, 1983, 35A: 100 -118. [19] FROGNER I L, HAAKENSTAD H, IVERSEN T. Limited-area ensemble predictions at the Norwegian Meteorological Institute[J]. Quart J R Meteor Soc, 2006, 132(621): 2 785-2 808. [20] ZOBEL Z, WANG J, WUEBBLES D J, et al. High-resolution dynamical downscaling ensemble projections of future extreme temperature distributions for the United States[J]. Earth\"s Future, 2017, 5(12): 1 234-1 251. [21] 马旭林, 计燕霞, 周勃旸, 等. GRAPES区域集合预报尺度混合初始扰动构造的新方案[J]. 大气科学学报, 2018, 41(2): 248-257. [22] LI J H, WAN Q L, GAO Y D, et al. The effect of sample optimization on the ensemble Kalman Filter in forecasting typhoon rammasun (2014)[J]. J Trop Meteor, 2018, 24(4): 433-447 [23] EVENSEN G. Sequential data assimilation with a nonlinear quasi-geostrophic model using Monte Carlo methods to forecast error statistics [J]. J Geophys Res, 1994, 99(C5): 10 143-10 162. [24] HOUTEKAMER P L, DEROME J. Data assimilation using an ensemble Kalman filter technique[J]. Mon Wea Rev, 1998, 126(3): 796-811. [25] WEI M, TOTH Z, WOBUS R, et al. Initial perturbations based on the ensemble transform (ET) technique in the NCEP global operational forecast system[J]. Tell, 2008(60A): 62-79. [26] TOTH Z, KALNAY E. Ensemble forecasting at NMC: The generationof perturbations[J]. Bull Amer Meteor Soc, 1993, 74(12): 2 317-2 330. [27] EHRENDORFER M, ERRICO R M, RAEDER K D. Singular-vector perturbation growth in a primitive equation model with moist physics [J]. J Atmos Sci, 1999, 56(11): 1627-1648. [28] PALMER T N, GELARO R, BARKMEIJER J, et al. Singular vectors, metrics, and adaptive observations[J]. J Atmos Sci, 1998, 55(4): 633- 653. [29] 杜钧, 陈静. 单一值预报向概率预报转变的基础: 谈谈集合预报及其带来的变革[J]. 气象, 2010, 36(11): 1-11. [30] TALAGRAND O, VAUTARD R. Evaluation of Probabilistic Prediction System[R]. Workshop on Predictability ECMWF, 1997(10): 20-22. [31] JOHN A S. The Relative Operating Characteristic in Psychology[J]. Science, 1973, 182(4 116): 990-1 000. [32] 谭燕, 陈德辉. 基于非静力模式物理扰动的中尺度集合预报试验[J]. 应用气象学报, 2007(3): 396-406+418. [33] WEN Y F, LIU Y D, TAN W C, et al. The Impact of Horizontal Resolution on the Intensity Andmicrostructure of Super Typhoon Usagi[J]. J Trop Meteor, 2019, 25(1): 24-33. -
点击查看大图
图(13) / 表(5)
计量
- 文章访问数: 191
- HTML全文浏览量: 18
- PDF下载量: 27
- 被引次数: 0
粤公网安备 4401069904700003号